CN204948075U - A kind of visible light communication receiving system based on neuroid - Google Patents

Abstract

The utility model discloses a kind of visible light communication method of reseptance based on neuroid, comprise the following steps: the vision signal that 1, information source sends by the first programmable gate array is converted to digital signal; 2, described digital signal is by LED drive circuit driving LED light fixture utilizing emitted light signal; 3, described light signal enters receiving subsystem through n photoelectricity testing part; 4, n described data flow is passed to the neuron after training and merges in module; 5, a described n data flow is made equilibrium treatment and exports merging data stream by described neuron merging module; Described merging data flows through the second programmable gate array demodulation process and is converted into vision signal.The invention also discloses a kind of visible light communication receiving system realizing the described visible light communication method of reseptance based on neuroid, comprising: launch subsystem, transmission subsystem and receiving subsystem.The advantages such as the utility model has broad application prospects.

Description

A kind of visible light communication receiving system based on neuroid

Technical field

The utility model relates to visible light communication technology, particularly a kind of visible light communication receiving system based on neuroid.

Background technology

Infraredly to compare with radio communication with traditional, LED visible light communication technology have transmitting power high, without electromagnetic interference, without the need to applying for the advantages such as the confidentiality of frequency spectrum resource and information.But visible light communication still faces many challenges, wherein one of ultimate challenge is that intersymbol interference greatly limit visible light communication system message transmission rate.White light LEDs for throwing light on is generally LED array form, and different point-source of light LED correspond to different light paths, and between light path, the delay of Signal transmissions can produce intersymbol interference; Simultaneously when system data transmission rate is higher, due to the restriction of LED bandwidth, the impact of a signal can be caused to expand on adjacent signals, produce intersymbol interference and error rate of system is promoted greatly; And because channel is undesirable, code element can be widened and time delay in process of transmitting, waveform shows as symbol pulses and produces hangover, the hangover of adjacent pulse can overlap each other, also can produce intersymbol interference, improve the error rate, thus have impact on the quality of communication.Generally, by changing coded system, as: non-return-to-zero coding in OOK modulation system is changed into NRZ or use orthogonal frequency division multiplexi to reduce intersymbol interference.But but substantially increase the complexity of visible light communication system.

In the communications fields such as scattering, relaying, microwave, the normal diversity receiving technology that uses solves the multipath effect caused due to the unsteadiness of channel parameter.Diversity receiving technology be same signal by mode distributed transmission such as different paths, time, angle, frequencies, receiving terminal obtains and organizes independent signal more, and by suitable merging mode, as selection combining, equal gain combining or maximum ratio addition etc.; Be merged into total Received signal strength by organizing independent signal more.Simultaneously in diversity system, because receiver module has multiple photodetector, be equivalent to the receiving area increasing photodiode, thus provide diversity gain for system, therefore diversity receiving technology can be used to improve the performance of visible light communication system.But due to time variation and the randomness of wireless channel, merging mode selected under different environment is different; And the data received are the initial data be disturbed in channel, not to the further optimization process of data.

Utility model content

In order to overcome the above-mentioned shortcoming of prior art with not enough, primary and foremost purpose of the present utility model is to provide a kind of visible light communication method of reseptance based on neuroid, this visible light communication method of reseptance optimizes VLC system channel performance further, do not increasing under bandwidth of a device prerequisite, significantly improving quality and the message transmission rate of radio communication.

In order to overcome the above-mentioned shortcoming of prior art with not enough, another object of the present utility model is the visible light communication receiving system of the visible light communication method of reseptance provided based on neuroid, this visible light communication receiving system is based on simple baseband modulation technique, by using angle diversity receiving technology to weaken the impact of intersymbol interference on system, and artificial neural network is used to merge the data that diversity reception obtains and optimize the error rate to reduce system.Overcome the diversity that traditional diversity receiving technology merges mode under various circumstances.

Primary and foremost purpose of the present utility model is achieved through the following technical solutions: a kind of visible light communication method of reseptance based on neuroid, comprises the following steps:

The vision signal that information source sends by step 1, the first programmable gate array is converted to digital signal;

Step 2, described digital signal are by LED drive circuit driving LED light fixture utilizing emitted light signal;

Step 3, described light signal enter receiving subsystem through n photoelectricity testing part, and wherein, n is positive integer; The n sub-channels of described n the corresponding transmission subsystem of photoelectric detector; N data flow of the corresponding receiving subsystem of described n sub-channels;

After step 4, a described n data flow through and amplify process, filtering process, be passed to the neuron after training and merge in module;

Step 5, described neuron merge module and a described n data flow are made equilibrium treatment and exports merging data stream; Described merging data flows through the second programmable gate array demodulation process and is converted into vision signal.

The neuron after training in described step 5 merges module and comprises the following steps:

When step 41, training, give described neuroid one group of input value and the desired value that matches with it, organize training data by this and adjust connection weighted value, contrasted by the real output value of described desired value and forward-propagating and obtain error signal;

$Y_i=f({\mathrm{\Σ}}_{j=1}^n{\mathrm{\ω}}_{ij}{V}_j-Q_i)$

E(x)＝[y(x)-F(x)] ²；

Step 42, utilize gradient descent method to carry out error back propagation and weights to correct, by repetition learning, described error signal is reached or lower than set point;

${\mathrm{\ω}}_{ij}(n+1)={\mathrm{\ω}}_{ij}\left(n\right)-\mathrm{\γ}\frac{\∂E\left(x\right)}{\∂{\mathrm{\ω}}_{ij}\left(n\right)};$

Step 43, described connection weighted value can be made to adjust to by the training of above-mentioned equation to make the mean square error between the actual output of neural net and described desired output minimum; After BP Algorithm for Training, for any input value, neuron equalizer all can provide comparatively suitable output, to complete the equilibrium treatment to channel.

Another object of the present utility model can be achieved through the following technical solutions: a kind of visible light communication receiving system realizing the described visible light communication method of reseptance based on neuroid, comprise: launch subsystem, transmission subsystem and receiving subsystem, described transmitting subsystem has: the first programmable gate array, LED drive circuit and LED lamp, and described receiving subsystem has: n photoelectricity testing part, ADC analog to digital converter, neuron merge module and the second programmable gate array; Described transmission subsystem is used for the visible light signal that LED lamp sends to be transferred to n photoelectricity testing part; The n sub-channels of described n the corresponding transmission subsystem of photoelectricity testing part; Described first programmable gate array, LED drive circuit, LED lamp, photoelectricity testing part, ADC analog to digital converter, neuron merge module and are connected successively with the second programmable gate array; Vision signal is converted to digital signal by described first programmable gate array, and described digital data transmission to LED drive circuit; Described LED drive circuit driving LED light fixture utilizing emitted light signal; Described light signal enters photoelectricity testing part through free space; A described n photoelectricity testing part converts analog signal to light signal, forms n circuit-switched data stream; Described ADC digital to analog converter samples into n railway digital signal the analog signal that photoelectricity testing part exports; Described n railway digital signal merges module through neuron to carry out merging and after equilibrium treatment, being passed to the second programmable gate array; Described second programmable gate array is reduced to vision signal signal through demodulation.

Described LED drive circuit comprises: information source, variable resistor, high-speed buffer, BiasTee module, DC current source and current-limiting resistance, and described information source, variable resistor, high-speed buffer, BiasTee module are connected successively with current-limiting resistance; The positive pole of described DC current source is connected with information source, the negative pole of described DC current source and BiasTee model calling; Described BiasTee module comprises electric capacity and inductance; One end of described inductance is connected with the negative pole of DC current source, and the other end of described inductance is connected with the negative pole of electric capacity, and the positive pole of described electric capacity is connected with high-speed buffer; The signal of telecommunication that described information source exports is transferred to BiasTee module through high-speed buffer, and the signal that the DC signal that described DC current source exports and high-speed buffer transmit is coupled in BiasTee module, generates coupled electrical signal; Described coupled electrical signal outputs to LED lamp by current-limiting resistance.

Described neuron merges its merging treatment of module by completing in programmable gate array signal processing chip or digital signal processing chip.

Described transmitting subsystem also comprises: the first liquid crystal display and video camera, and described receiving subsystem also comprises amplifying circuit, filter circuit and the second liquid crystal display; The first described liquid crystal display is all connected with the first programmable gate array with video camera, and photoelectricity testing part is connected with ADC digital to analog converter with rearmounted amplifying circuit by pre-amplification circuit, and the second programmable gate array is connected with the second liquid crystal display; Video signal transmission is given the first programmable gate array by described video camera; Described first programmable gate array gives the first liquid crystal display video signal transmission; Described photoelectricity testing part carries out amplification process analog signal by amplifying circuit; Described analog signal after amplification after filtering circuit is carried out filtering process and is transferred to ADC analog to digital converter again; Described second programmable gate array gives the second liquid crystal display video signal transmission.

Another object of the present utility model also can be achieved through the following technical solutions: a kind of visible light communication receiving system of the visible light communication method of reseptance based on neuroid, comprise: launch subsystem, transmission subsystem and receiving subsystem, data input in LED lamp by LED drive circuit by described transmitting subsystem, send visible ray, by transmission subsystem, light signal is transmitted effectively, finally by receiving subsystem, the induction change of light intensity is realized communicating with the conversion of data.

Described transmitting subsystem is made up of video camera, the first programmable gate array, the first liquid crystal display, LED drive circuit, LED lamp, and further, described LED drive circuit is made up of the electricity component such as electric capacity and inductance.

Described transmission subsystem is free space, has n subchannel, and wherein, n is positive integer.

Described receiving subsystem is made up of n photoelectric detector, amplifying circuit, filter circuit, neuron merging module, the second programmable gate array, the second liquid crystal display.

Principle of the present utility model: the utility model utilance diversity receiving technology and neuron merge network, described diversity receiving technology is that same signal is by mode distributed transmission such as different paths, time, angle, frequencies, organize independent signal at receiving terminal by multiple photoelectricity testing part acquisition more, and by suitable merging mode, as selection combining, equal gain combining or maximum ratio addition etc.; Be merged into total Received signal strength by organizing independent signal more.Simultaneously because diversity receiving technology has multiple photoelectric detector, be equivalent to the effective area increasing receiver light-receiving, therefore increase received signal to noise ratio when the signal to noise ratio of information source is constant, thus make that error rate of system is lower, fault-tolerance is stronger.It is carry out merging optimizing to obtained multi-group data forming total output signal that described neuron merges module.By the Nonlinear Mapping relation using nonlinear transfer function to realize input and output.And then filtering process, estimation channel being carried out to phase shift and compensation etc. are carried out to realize equilibrium treatment to the signal be interfered.Due to the impact of intersymbol interference, the error rate of visible light communication system is caused to promote, for this reason, by using diversity receiving technology to weaken the impact of intersymbol interference on system, and artificial neural network is used to merge the data that diversity reception obtains and optimize the error rate to reduce system.This receiving system can reduce the impact of intersymbol interference on system effectively, improves the signal to noise ratio of Received signal strength, reduces the error rate of system.Do not increasing under bandwidth of a device prerequisite, significantly improving quality and the message transmission rate of radio communication; Have broad application prospects.

Compared with prior art, the utility model has the following advantages and beneficial effect:

1, the utility model is based on simple baseband modulation technique, compared with the traditional multiplexing modulation of use orthogonal frequency division, discrete multitone techniques etc., enormously simplify the complexity of system, and under the prerequisite of LED component bandwidth can not be increased, improve the quality of radio communication and the capacity of channel exponentially.

2, the utility model has used artificial neural network to carry out merging optimization to the multi-group data that diversity receiving technology obtains, and while achieving the impact eliminating intersymbol interference, reduces the error rate, serves the effect of post-equalization.

Accompanying drawing explanation

Fig. 1 is the schematic diagram that the utility model realizes based on the visible light communication receiving system of neuroid.

Fig. 2 is the schematic diagram of the utility model LED drive circuit.

Fig. 3 is the utility model MLP neuroid schematic diagram.

Fig. 4 position the utility model uses the schematic diagram of BP training algorithm.

Fig. 5 is the operating diagram that the utility model neuron merges module.

Embodiment

Below in conjunction with embodiment and accompanying drawing, the utility model is described in further detail, but execution mode of the present utility model is not limited thereto.

Embodiment

As shown in Figure 1, a kind of visible light communication system based on neuroid.Comprise: launch subsystem, transmission subsystem, receiving subsystem.Described transmitting subsystem, video camera receives real time video signals and is transferred to the first liquid crystal display by the first programmable gate array; Described first liquid crystal display demonstrates raw video signal; Described first programmable gate array is modulated described real time video signals and after corresponding video signal processing technology, is transferred in LED drive circuit; Visible light signal launched by described LED drive circuit driving LED light fixture.As shown in Figure 2, be the schematic diagram of LED drive circuit, closed by the electric capacity of BiasTee structure and inductance T junction, realize being coupled thus ensure that signal can not be lost in the led of direct current signal and AC signal.

Further, light signal arrives receiving subsystem by transmission subsystem; Described transmission subsystem is 4 different channels in the present embodiment.Described light signal is converted to the signal of telecommunication through 4 photoelectric detectors respectively, form 4 circuit-switched data stream; 4 described circuit-switched data streams carry out filter amplifying processing by amplifying circuit and filter circuit, then carry out quantification by ADC analog to digital converter to the signal of telecommunication and be converted to digital signal.Then merge module by neuron and merging treatment acquisition final data stream is carried out to 4 circuit-switched data streams.

Further, described neuron merges the MLP neuroid that module uses three layers, and uses BP algorithm to carry out learning training to neuroid.As shown in Figure 3, be the structural representation of three layers of MLP neuroid; As shown in Figure 4, be BP training algorithm.For i-th neuronic output state Y _ifor:

$Y_i=f({\mathrm{\Σ}}_{j=1}^n{\mathrm{\ω}}_{ij}{V}_j-Q_i)$

In formula, to be i-th neuron neuronic with jth is connected weighting value; V _jit is a jth neuronic output; Q _ii-th neuronic threshold value; For activation primitive.

Give described neuroid one group of input value and the desired value that matches with it during training, adjust connection weighted value by this group training data.Contrast can obtain error signal by the output valve of described desired value and forward-propagating.

E(x)＝[y(x)-F(x)] ²，

Wherein, x and y (x) real output signal that represents input signal respectively and correspond; F (x) is for expecting output signal.Utilize gradient descent method to carry out error back propagation and weights to correct, by repetition learning, described error signal is reached or lower than set point.

${\mathrm{\ω}}_{ij}(n+1)={\mathrm{\ω}}_{ij}\left(n\right)-\mathrm{\γ}\frac{\∂E\left(x\right)}{\∂{\mathrm{\ω}}_{ij}\left(n\right)};$

Wherein, ω _ijrepresentative connects weighted value, and γ represents learning rate.Described connection weighted value can be made to adjust to by the training of above-mentioned equation makes the mean square error between the actual output of neural net and described desired output minimum.After BP Algorithm for Training, for any input value, neuron equalizer all can provide comparatively suitable output.And then the equilibrium treatment achieved channel.As shown in Figure 5, n circuit-switched data is merged into a circuit-switched data stream by the neuroid after training, and to arrange a threshold value at neuronic output be 0.5 to produce binary code.Wherein, in this example, n=4.

Signal is converted to vision signal by the second programmable gate array by described final data stream again, then passes in the second liquid crystal display.Compared with signal in the second liquid crystal display by the first liquid crystal display, the error rate of system can be tested out, for the operation of verification system.

Above-described embodiment is only a kind of execution mode of the present utility model; but execution mode of the present utility model is not restricted to the described embodiments; other are any do not deviate from Spirit Essence of the present utility model and principle under do change, modification, substitute, combine, simplify the substitute mode that all should be equivalence, be included within protection range of the present utility model.

Claims (3)

1. the visible light communication receiving system based on neuroid, comprise: launch subsystem, transmission subsystem and receiving subsystem, it is characterized in that, described transmitting subsystem has: the first programmable gate array, LED drive circuit and LED lamp, and described receiving subsystem has: n photoelectricity testing part, ADC analog to digital converter, neuron merge module and the second programmable gate array; Described first programmable gate array, LED drive circuit, LED lamp, photoelectricity testing part, ADC analog to digital converter, neuron merge module and are connected successively with the second programmable gate array.

2. the visible light communication receiving system based on neuroid according to claim 1, it is characterized in that, described LED drive circuit comprises: information source, variable resistor, high-speed buffer, BiasTee module, DC current source and current-limiting resistance, and described information source, variable resistor, high-speed buffer, BiasTee module are connected successively with current-limiting resistance; The positive pole of described DC current source is connected with information source, the negative pole of described DC current source and BiasTee model calling; Described BiasTee module comprises electric capacity and inductance; One end of described inductance is connected with the negative pole of DC current source, and the other end of described inductance is connected with the negative pole of electric capacity, and the positive pole of described electric capacity is connected with high-speed buffer.

3. the visible light communication receiving system based on neuroid according to claim 1, it is characterized in that, described transmitting subsystem also comprises: the first liquid crystal display and video camera, and described receiving subsystem also comprises amplifying circuit, filter circuit and the second liquid crystal display; The first described liquid crystal display is all connected with the first programmable gate array with video camera, and photoelectricity testing part is connected with ADC digital to analog converter with rearmounted amplifying circuit by pre-amplification circuit, and the second programmable gate array is connected with the second liquid crystal display.