Background technique
Cognitive radio (cognitive radio, CR) is left unused frequency spectrum, is passed through by perception external environment, dynamic access
Dynamic Spectrum Management mechanism realizes primary and secondary user sharing frequency spectrum using spectrum interposition.
Vehicle-mounted short haul connection (Vehicle to X:V2X) network be by wireless communication, the short distances such as GPS/GIS, sensing
Car (CAN-Controller Area Network), bus or train route (Vehicle-2-RSU), workshop from communication technology realization
(Vehicle-2-Vehicle), between vehicle outer (vehicle-2-Infrastructure), people's vehicle (Vehicle-2-Person)
Communication.
Frequency range of the V2X with fixed authorization, but narrower bandwidth.Bulk information in V2X services class extension class business, by
In the limitation of authorized bandwidth, the expansion of its type of business and the diversity of business tine are influenced.
Cognitive radio technology can effectively expand the working frequency range of V2X, provide the channel of high bandwidth for extension class business,
Since extension class of the V2X based on information service be not stringent to delay requirement, it is highly suitable for transmitting in cognition channel.
Cognitive system in V2X is that a kind of dynamic utilizes spectrum system, from the perspective of single user, dynamic
The frequency spectrum usage behavior being embodied between cognitive user influences each other.Power control is an important topic in communication network,
The purpose is to inhibit network interferences, save the energy of terminal device and guarantee the service quality of communication link.For recognizing net
Network, power control can also control the interference to authorized user, thus the optimization of power control cognitive system is played it is important
Effect.
Existing power control algorithm is largely the communication network fixed for communication node.Due to the vehicle-mounted section in V2X
In fast moving, the variation of the distance between node and node of surrounding frequently, cannot be saved directly using existing cognition point
Point power control optimization algorithm.In addition, many detection nodes that are arranged think cognitive nodes in part cognition power control optimization algorithm
Cognitive nodes work is fed back to the interference temperature of primary user, in the vehicle that V2X is fast moved, establishes this kind of dedicated detection node
Also relatively difficult.
Summary of the invention
The present invention provides a kind of wireless cognition channel power distribution optimizations suitable for vehicle-mounted short distance communication network to calculate
Method, including three parts.First part is region Interference Detection model foundation, by authorization channel periodic broadcasting comprising vehicle-mounted
The unique identification of node and the heartbeat message in geographical location obtain the information of the vehicle-mounted node of surrounding, then pass through cognition work
Node state broadcast obtains the Node distribution and working channel of one's respective area cognition working node.Other do not operate at cognition letter
The node in road is arranged to cognition detection node, and the signal sent by receiving cognition working node calculates cognitive interference temperature;The
Two parts are that parameter obtains and power distribution optimization algorithm, vehicle-mounted node obtain cognition channel and correspondence by trackside node broadcasts
Jamming power thresholding, then determine Optimized model using lagrangian optimization algorithm and game theory;Part III is cognition
Working node power optimization is obtained interference temperature by detection node, is then being authorized using the power distribution optimization algorithm of design
Multi-band broadcast gives cognition working node, and cognition working node is adjusted the power of this node using the algorithm of setting, repeats
The above process obtains the optimization of power distribution in one's respective area.
The present invention provides a kind of wireless cognition channel powers suitable for vehicle-mounted short distance communication network to distribute optimization side
Method, which is characterized in that vehicle-mounted node obtains cognition channel and corresponding jamming power thresholding by trackside node broadcasts, by fixed
The message comprising node unique identification and geographical location of phase broadcast, obtains the information of the vehicle-mounted node of surrounding.Recognize work section
Point broadcasts its state, other nodes are arranged to cognition detection node, calculate cognitive interference, then utilizes lagrangian optimization algorithm
With game theory algorithm, interference temperature is obtained by detection node, then cognition working node is broadcast in authorized spectrum band, recognizes work
Node is adjusted the power of this node, repeats the above process, and obtains the optimization of power distribution in one's respective area.
Preferably, specifically comprise the following steps S1, S2, S3:
S1, region Interference Detection model foundation comprising
S11, cognition working node state broadcast;
The vehicle-mounted node mode setting of S12, cognitive interference detecting state;
The power distribution Optimized model of S13, vehicle-mounted node;
S2, parameter obtains and power distribution optimization algorithm comprising
S21, detection node jamming power threshold sets;
S22, Lagrange operator and relaxation;
S23, vector λ are solved;
S24, Best response function solve;
S3, cognition working node power optimization comprising
S31, detection node broadcast interference temperature;
S32, cognitive nodes adjust power;
S33, variable update;
S34, circulation.
Preferably, it in the step S11, is recognized using the vehicle-mounted node of cognition multi-channel operation in authorization channel periodic broadcasting
The channel use state of channel, so that cognition detection node can know that the cognition channel that needs detect;
Preferably, in the step S12, the vehicle-mounted node of cognition channel is not operated at as cognition detection node, is mentioned
For recognizing Interference Detection;
Preferably, it in the step S13, is limited by utility of power function and interference temperature to construct power distribution optimization
Model;
Preferably, in the step S21, trackside node provides cognition channel and corresponding jamming power thresholding, passes through authorization
Channels broadcast is saved in local after giving vehicle-mounted node, vehicle-mounted node to receive, and receives in next trackside node new similar
Local information is updated after type broadcast;
Preferably, in the step S31, after cognition detection node calculates interference temperature, it is broadcast to cognition working node, is recognized
Working node is known according to the interference temperature of each detection node feedback received, and adjustment signal sends power, to optimize local area
The power distribution of cognition working node in domain.
The present invention utilizes the vehicle-mounted node in regional area, not using cognition channel in V2X network to undertake detection interference
Then the work of temperature feeds back to cognition working node in authorization channel, to realize the power distribution optimization of whole region.
Specific embodiment
The present invention provides a kind of wireless cognition channel powers suitable for vehicle-mounted short distance communication network to distribute optimization side
Method, this method include three parts in step, are region Interference Detection model foundation respectively, parameter obtains and power distribution optimization
Algorithm, cognition working node power optimization, the present invention is described in further detail below.
First part, region Interference Detection model foundation
In regional area, vehicle-mounted node is in the broadcast of authorized spectrum band periodically comprising vehicle unique identification (VIN) and geographical
The security classes applications such as message (abbreviation heartbeat message) user's anticollision of position.Vehicle-mounted node is wide by vehicle-mounted node around receiving
The heartbeat message broadcast can obtain the quantity and respective positions of the vehicle-mounted node of coverage area inner periphery.If in some region in Q,
There are T vehicle-mounted nodes that can mutually receive respective heartbeat message.
1.1, cognition working node state broadcast
It is located in the region Q, has N number of vehicle-mounted node work in cognition channel, these nodes of default are in authorization channel
Periodicity broadcasts the working condition of the cognition channel of its occupancy, and format is as shown in Figure 2, in which:
Type: Packet type, mark are the notification broadcasts that vehicle-mounted node occupies cognition channel;
VIN: the unique identification of vehicle-mounted node, the different vehicle-mounted nodes of difference;
GP: the geographical location of vehicle-mounted node is derived from the GPS or Beidou equipment of vehicle-mounted node;
Ch-XML: it using the service condition of the XML vehicle-mounted node cognition channel of sheet described, notifies in coverage area (general
150 meters of radius) the cognition channels description currently in use of other vehicle-mounted this nodes of node, including channel frequency, bandwidth and work
Make mode.
1.2, the vehicle-mounted node mode setting of cognitive interference detecting state
It is located in the Q of region, there are M=T-N vehicle-mounted nodes not operate at cognition frequency range, the work of these nodes of default
In interference measurement operating mode, interference measurement is carried out using its idle cognition module.By receiving broadcast in authorized spectrum band
Channel channels broadcast packet is recognized, obtains work in the quantity of the vehicle-mounted node of cognition channel and the occupancy situation of cognition channel.
1.3, the power distribution Optimized model of vehicle-mounted node
In the Q of region, vehicle-mounted node interference temperature measurement model is established, as shown in figure 3, comprising N to receiving and dispatching vehicle-mounted node,
Interference level of the vehicle-mounted node monitoring cognitive nodes of M interference measurement to authorized user.
To any one node i in N number of cognition working node, letter drying is more as shown in Equation 1 than SINR:
Wherein, i={ 1,2 ..., N }, PiPower, G are sent for itiFor its channel gain, σ2For noise.
To the arbitrary node k in M detection node, to guarantee not interfere primary user, received total jamming power should expire
Sufficient formula 2:
Wherein, k={ 1,2 ..., M }, gikBe recognize working node i to detection node k channel gain, B be Bohr hereby
Graceful constant, TkIt is interference temperature thresholding, is that maximum interference allowed by authorized user receiving end determines by default.
The utility function that working node i is recognized in definition region Q is ui(pi)=log (SINRi), channel function is recognized for it
The optimum allocation of the function of rate, each cognitive nodes power can be described with formula 3:
Subject to
Subject to is " being limited to ", illustrates that the optimal power allocation of (formula 3) is limited to each test point of (formula 4)
The limitation that power cannot interfere primary user to work.
Second part, parameter obtains and power distribution optimization algorithm
Due to vehicle be in movement in, obtain it is available cognition channel and its channel characteristic, need a region compared with
Prolonged perception, vehicle realize that this function is relatively difficult, and the present embodiment can use cognition channel by scanning by trackside node
And the characteristic of learning cognition channel, the frequency including channel, bandwidth, jamming power thresholding, broadcast packet is formed, broadcast distribution is given
The vehicle-mounted node passed through, vehicle-mounted node are used and are detected for cognition channel after saving and use, until next trackside node is issued
New data.
2.1, detection node jamming power threshold sets
Cognition detection node is according to the working frequencies for recognizing working nodes all in region in the Q of region, from local cache
Obtain corresponding jamming power thresholding Zk=BTk, illustrate the maximum interference power that detection node k allows.
2.2, Lagrange operator and relaxation
It is shown to obtain Lagrangian such as (formula 5) for the condition limitation that (formula 4) is released using Lagrange multiplier method:
Optimized power distribution is equivalent to solution (formula 6):
2.3, vector λ is solved
Vector λ={ λ1,λ2,...,λmAcquired by (formula 7):
λ (t+1)=max { 0, λk(t)+lk(t)(Nk(t)-Zk) (formula 7)
T indicates operation moment, and (t+1) indicates next operation moment, Nk(t) it indicates the general power that node k is detected, is equal to
lkIt indicates to update step-length, be acquired by (formula 9):
Wherein, α is a normal number of default.
2.4, Best response function solves
Use piIndicate the transmission power of cognition working node i, p-iIndicate other nodes in addition to recognizing working node i
Transmission power, then under the premise of other node powers are constant, the optimal power contribution of node i can be expressed as optimum response letter
Several solutions is acquired according to game theory algorithm by (formula 10):
Wherein, symbolIt indicates max { a, min (b, X) }, πjIt is the testing result of detection node j feedback.
Part III, cognition working node power optimization
3.1, detection node broadcasts interference temperature
Detection node k calculates the interference temperature detected, such as shown in (formula 11):
And by interference temperature πkIn authorized channels according to the time interval of default, periodic broadcast to region Q.
3.2, cognitive nodes adjust power
Cognitive nodes update real-time function after the interference temperature that authorization channel receives detection node broadcast, with (formula 12)
Rate:
3.3, variable update
Variable λ is updated using (formula 13) according to subgradient algorithmk, step-length lk
λk(t+1)=max { 0, λk(t)+lk(t)(Nk(t)-Zk) (formula 13)
Step-length is lkIt is acquired by (formula 9).
3.4, it recycles
T=t+1 goes to 3.1, and the power for executing next cycle updates, and each cognitive nodes of region Q obtain power point
The optimization matched.
The above is presently preferred embodiments of the present invention and its technical principle used, for those skilled in the art
For, without departing from the spirit and scope of the present invention, any equivalent change based on the basis of technical solution of the present invention
Change, simple replacement etc. is obvious changes, all fall within the protection scope of the present invention.