CN103248443A - Method for sensing OFDM spectrum under conditions of time asynchronization and known cyclic prefix length - Google Patents

Abstract

The invention discloses a method for sensing an OFDM (Orthogonal Frequency Division Multiplexing) spectrum under conditions of time asynchronization and known cyclic prefix length. The processing procedure of the method comprises the following steps: at first, sampling a receipt signal from a monitoring channel to obtain a sampled signal; then, computing an autocorrelation function of the sampled signal according to the sampled value in the sampled signal; next, computing test statistic according to an autocorrelation coefficient in the autocorrelation function of the sampled signal under the condition of time asynchronization; and at last, judging whether the monitoring channel is in the idle state or not according to the test statistic and the judgment threshold. The method has the advantages that the spectrum sensing can be carried out by directly using the sampled signal, without the need of time synchronization, so that the computation complexity is reduced effectively. During the process of computing the test statistic, due to the use of the nonnegative characteristic of the autocorrelation coefficient corresponding to the cyclic premix part of the OFDM signal in the autocorrelation function of the sampled signal, the spectrum sensing performance of the OFDM signal is improved effectively.

Description

OFDM frequency spectrum sensing method under the asynchronous and circulating prefix-length known case of time

Technical field

The present invention relates to the frequency spectrum perception technology in a kind of cognitive radio system, especially relate to the OFDM frequency spectrum sensing method under the asynchronous and circulating prefix-length known case of a kind of time.

Background technology

Along with the fast development of various radio communication services, importance and the scarcity of frequency spectrum resource highlight day by day.A large amount of studies show that, current frequency spectrum shortage present situation is not owing to frequency resource deficiency physically, and mainly the availability of frequency spectrum is low to be caused because fixed frequency spectrum allocation manager mechanism causes.Cognitive radio is attempted to propose from the resource problem more and more in short supply that solves radio communication in essence by improving the availability of frequency spectrum just.Its communication environment around can real-time perception, identify available idle channel, adjust the system parameters of cognition wireless network then according to the frequency spectrum perception result adaptively, make cognitive radio system have intelligent identification and the ability that changes frequency spectrum use chance.In order to prevent that existing communication system is produced interference, cognitive radio system must identify idle channel effectively reliably, so frequency spectrum perception is one of key technology in the cognitive radio.

OFDM (Orthogonal frequency division multiplexing, OFDM) technology has characteristics such as availability of frequency spectrum height, and this technology is technology current and that future, wireless communication standard was widely adopted.Therefore the frequency spectrum perception (namely judge and whether have ofdm signal in the channel) to ofdm signal has very important significance.Existing frequency spectrum sensing method at ofdm signal mainly can be divided into frequency domain detection method and time domain detection method two classes.Wherein, the frequency domain detection method need be calculated the frequency spectrum of gathering signal, therefore has bigger amount of calculation; The time domain detection method mainly utilizes the autocorrelation performance of Cyclic Prefix in the ofdm signal to realize frequency spectrum perception.People such as Chaudhari have proposed to utilize the autocorrelation performance of Cyclic Prefix in " in the Autocorrelation-Based Decentralized Sequential Detection of OFDM Signals in Cognitive Radios(cognitive radio based on autocorrelative ofdm signal distributed sequential detection method) " in 2009, realize frequency spectrum perception by calculating the auto-correlation function that receives signal, but this method is not considered the non-stationary property of auto-correlation function.At this problem, people such as Axell have proposed a kind of frequency spectrum sensing method of new auto-correlation function based on the reception signal in " the optimum and suboptimum frequency spectrum sensing method of the OFDM during the known and unknown noise variance of Optimal and Sub-Optimal Spectrum Sensing of OFDM Signals in Known and Unknown Noise Variance() " in 2011, the method that people such as this method and Chaudhari propose is compared, this method has more excellent detection performance, but this method requires time synchronized.

Summary of the invention

Technical problem to be solved by this invention provides the OFDM frequency spectrum sensing method under the asynchronous and circulating prefix-length known case of a kind of time, and it can improve the frequency spectrum perception performance of ofdm signal effectively, and computation complexity is low.

The present invention solves the problems of the technologies described above the technical scheme that adopts: the OFDM frequency spectrum sensing method under the asynchronous and circulating prefix-length known case of a kind of time, the processing procedure that it is characterized in that it is: at first, reception signal from supervisory channel is sampled, obtain sampled signal; Then, according to the sampled value in the sampled signal, the auto-correlation function of calculating sampling signal; Then, under the time asynchronous condition, according to the auto-correlation coefficient in the auto-correlation function of sampled signal, calculate test statistics; At last, according to the size of test statistics and decision threshold, judge whether supervisory channel is in idle condition.

It specifically may further comprise the steps:

1. utilize the sampling module in the cognitive radio system that the reception signal from supervisory channel is carried out M sampling, obtain the sampled signal that the sampled value by M sampled point constitutes, wherein, M=K * (N _c+ N _d)+N _d, K represents any positive integer, N _cThe length of the Cyclic Prefix of expression ofdm signal, N _dThe number of the subcarrier of expression ofdm signal;

2. according to the sampled value in the sampled signal, the auto-correlation function of calculating sampling signal is designated as ρ (t) with t auto-correlation coefficient in the auto-correlation function of sampled signal, $\mathrm{\ρ}\left(t\right)=\frac{\frac{1}{K}\underset{k=1}{\overset{K}{\mathrm{\Σ}}}x((k-1)\×(N_c+N_d)+t)x((k-1)\×(N_c+N_d)+N_d+t)}{\frac{1}{M}\underset{m=1}{\overset{M}{\mathrm{\Σ}}}{x}^2\left(m\right)},$ Wherein, the number of the auto-correlation coefficient in the auto-correlation function of sampled signal is N _c+ N _d, t is positive integer, and 1≤t≤N _c+ N _d, k is positive integer, and 1≤k≤K, x ((k-1) * (N _c+ N _d((k-1) * (N in the expression of the)+t) sampled signal _c+ N _dThe individual sampled value of)+t), x ((k-1) * (N _c+ N _d)+N _d+ t) ((k-1) * (N of expression in the sampled signal _c+ N _d)+N _d+ t) individual sampled value, m is positive integer, and 1≤m≤M, m sampled value in x (m) the expression sampled signal;

3. under the time asynchronous condition, according to the auto-correlation coefficient in the auto-correlation function of sampled signal, calculate test statistics, be designated as T, $T=\frac{1}{N_c+N_d}\underset{t=1}{\overset{N_c+N_d}{\mathrm{\Σ}}}\mathrm{\ρ}\left(t\right)\×\left|\mathrm{\ρ}\left(t\right)\right|,$ Wherein, symbol " || " is the symbol that takes absolute value;

4. test statistics T and decision threshold λ are compared, if T, judges then that supervisory channel is in busy condition greater than λ, if T is less than or equal to λ, judge that then supervisory channel is in idle condition, wherein, λ=[F ^-1(P _f)] ², P _fThe expression false alarm probability, span is 0＜P _f＜0.5, F ^-1() is the inverse function of F (),

$F\left(\mathrm{\λ}\right)=\frac{\sqrt{K(N_c+N_d)}}{\sqrt{\mathrm{\π}}}{\∫}_{\mathrm{\λ}}^{+\∞}{e}^{-K(N_c+N_d)y^2}\mathrm{dy},$ E is the nature radix, e=2.71828 ..., y is variable.

Compared with prior art, the invention has the advantages that:

1) the inventive method since when not required synchronously, directly utilize sampled signal to carry out frequency spectrum perception, therefore reduced computation complexity effectively.

2) the inventive method is in the process of calculating test statistics, owing to the non-negative characteristic that auto-correlation coefficient corresponding with the Cyclic Prefix part of ofdm signal in the auto-correlation function that has utilized sampled signal has, therefore improved the frequency spectrum perception performance of ofdm signal effectively.

Description of drawings

Fig. 1 is the FB(flow block) of frequency spectrum sensing method of the present invention;

Fig. 2 is under different state of signal-to-noise, and the frequency spectrum sensing method that people such as Axell propose and the detection probability of the inventive method be schematic diagram relatively.

Embodiment

Describe in further detail below in conjunction with the present invention of accompanying drawing embodiment.

OFDM frequency spectrum sensing method under the asynchronous and circulating prefix-length known case of a kind of time that the present invention proposes, its FB(flow block) as shown in Figure 1, its main processing procedure is: at first, the reception signal from supervisory channel is sampled, obtain sampled signal; Then, according to the sampled value in the sampled signal, the auto-correlation function of calculating sampling signal; Then, under the time asynchronous condition, according to the auto-correlation coefficient in the auto-correlation function of sampled signal, calculate test statistics; At last, according to the size of test statistics and decision threshold, judge whether supervisory channel is in idle condition.

Ofdm signal frequency spectrum sensing method of the present invention, it specifically may further comprise the steps:

1. utilize the sampling module in the cognitive radio system that the reception signal from supervisory channel is carried out M sampling, obtain the sampled signal that the sampled value by M sampled point constitutes, wherein, M=K * (N _c+ N _d)+N _d, K represents any positive integer, as gets K=10, N _cThe length of the Cyclic Prefix of expression ofdm signal, N _dThe number of the subcarrier of expression ofdm signal.

2. according to the sampled value in the sampled signal, the auto-correlation function of calculating sampling signal is designated as ρ (t) with t auto-correlation coefficient in the auto-correlation function of sampled signal, $\mathrm{\ρ}\left(t\right)=\frac{\frac{1}{K}\underset{k=1}{\overset{K}{\mathrm{\Σ}}}x((k-1)\×(N_c+N_d)+t)x((k-1)\×(N_c+N_d)+N_d+t)}{\frac{1}{M}\underset{m=1}{\overset{M}{\mathrm{\Σ}}}{x}^2\left(m\right)},$ Wherein, the number of the auto-correlation coefficient in the auto-correlation function of sampled signal is N _c+ N _d, t is positive integer, and 1≤t≤N _c+ N _d, k is positive integer, and 1≤k≤K, x ((k-1) * (N _c+ N _d((k-1) * (N in the expression of the)+t) sampled signal _c+ N _dThe individual sampled value of)+t), x ((k-1) * (N _c+ N _d)+N _d+ t) ((k-1) * (N of expression in the sampled signal _c+ N _d)+N _d+ t) individual sampled value, m is positive integer, and 1≤m≤M, m sampled value in x (m) the expression sampled signal.

3. under the time asynchronous condition, according to the auto-correlation coefficient in the auto-correlation function of sampled signal, calculate test statistics, be designated as T, $T=\frac{1}{N_c+N_d}\underset{t=1}{\overset{N_c+N_d}{\mathrm{\Σ}}}\mathrm{\ρ}\left(t\right)\×\left|\mathrm{\ρ}\left(t\right)\right|,$ Wherein, symbol " || " is the symbol that takes absolute value.

4. test statistics T and decision threshold λ are compared, if T, judges then that supervisory channel is in busy condition greater than λ, if T is less than or equal to λ, judge that then supervisory channel is in idle condition, wherein, λ=[F ^-1(P _f)] ², P _fThe expression false alarm probability, span is 0＜P _f＜0.5, F ^-1() is the inverse function of F (), $F\left(\mathrm{\λ}\right)=\frac{\sqrt{K(N_c+N_d)}}{\sqrt{\mathrm{\π}}}{\∫}_{\mathrm{\λ}}^{+\∞}{e}^{-K(N_c+N_d)y^2}\mathrm{dy},$ E is the nature radix, e=2.71828 ..., y is variable.

By feasibility and the validity of following emulation to further specify frequency spectrum sensing method of the present invention.

The number of supposing the subcarrier of ofdm signal is N _d=32, the length of the Cyclic Prefix of ofdm signal is N _c=8, get K=10, then total sampling number is M=432, and is P according to the value that false alarm probability is established in the requirement of IEEE802.22 standard _f=0.1.Fig. 2 provided signal to noise ratio from-when 20dB changes to 5dB, the frequency spectrum sensing method that people such as Axell propose and the comparison of the inventive method by the detection probability of 100000 Monte Carlo simulation acquisitions.As can be seen from Figure 2, the detection probability of the inventive method is far superior to the detection probability of the frequency spectrum sensing method that people such as Axell proposes.Analysis chart 2, when signal to noise ratio be-during 1dB, the detection probability of the inventive method is 0.94, reached detection probability in the IEEE802.22 standard more than or equal to 0.9 requirement, and this moment the frequency spectrum sensing method that people such as Axell proposes detection probability can only reach 0.8, this does not reach detection probability in the IEEE802.22 standard more than or equal to 0.9 requirement; When signal to noise ratio is 1dB, the detection probability 0.93 of the frequency spectrum sensing method that people such as Axell propose, and the detection probability of the inventive method is near 1 at this moment, this is enough to illustrate feasibility and the validity of the inventive method.

Claims (2)

1. the OFDM frequency spectrum sensing method under the asynchronous and circulating prefix-length known case of time is characterized in that its processing procedure is: at first, the reception signal from supervisory channel is sampled, obtain sampled signal; Then, according to the sampled value in the sampled signal, the auto-correlation function of calculating sampling signal; Then, under the time asynchronous condition, according to the auto-correlation coefficient in the auto-correlation function of sampled signal, calculate test statistics; At last, according to the size of test statistics and decision threshold, judge whether supervisory channel is in idle condition.

2. the OFDM frequency spectrum sensing method under the asynchronous and circulating prefix-length known case of time according to claim 1 is characterized in that it specifically may further comprise the steps:

1. utilize the sampling module in the cognitive radio system that the reception signal from supervisory channel is carried out M sampling, obtain the sampled signal that the sampled value by M sampled point constitutes, wherein, M=K * (N _c+ N _d)+N _d, K represents any positive integer, N _cThe length of the Cyclic Prefix of expression ofdm signal, N _dThe number of the subcarrier of expression ofdm signal;

2. according to the sampled value in the sampled signal, the auto-correlation function of calculating sampling signal is designated as ρ (t) with t auto-correlation coefficient in the auto-correlation function of sampled signal, $\mathrm{\ρ}\left(t\right)=\frac{\frac{1}{K}\underset{k=1}{\overset{K}{\mathrm{\Σ}}}x((k-1)\×(N_c+N_d)+t)x((k-1)\×(N_c+N_d)+N_d+t)}{\frac{1}{M}\underset{m=1}{\overset{M}{\mathrm{\Σ}}}{x}^2\left(m\right)},$ Wherein, the number of the auto-correlation coefficient in the auto-correlation function of sampled signal is N _c+ N _d, t is positive integer, and 1≤t≤N _c+ N _d, k is positive integer, and 1≤k≤K, x ((k-1) * (N _c+ N _d((k-1) * (N in the expression of the)+t) sampled signal _c+ N _dThe individual sampled value of)+t), x ((k-1) * (N _c+ N _d)+N _d+ t) ((k-1) * (N of expression in the sampled signal _c+ N _d)+N _d+ t) individual sampled value, m is positive integer, and 1≤m≤M, m sampled value in x (m) the expression sampled signal;

3. under the time asynchronous condition, according to the auto-correlation coefficient in the auto-correlation function of sampled signal, calculate test statistics, be designated as T, $T=\frac{1}{N_c+N_d}\underset{t=1}{\overset{N_c+N_d}{\mathrm{\Σ}}}\mathrm{\ρ}\left(t\right)\×\left|\mathrm{\ρ}\left(t\right)\right|,$ Wherein, symbol " || " is the symbol that takes absolute value;

4. test statistics T and decision threshold λ are compared, if T, judges then that supervisory channel is in busy condition greater than λ, if T is less than or equal to λ, judge that then supervisory channel is in idle condition, wherein, λ=[F ^-1(P _f)] ², P _fThe expression false alarm probability, span is 0＜P _f＜0.5, F ^-1() is the inverse function of F (), $F\left(\mathrm{\λ}\right)=\frac{\sqrt{K(N_c+N_d)}}{\sqrt{\mathrm{\π}}}{\∫}_{\mathrm{\λ}}^{+\∞}{e}^{-K(N_c+N_d)y^2}\mathrm{dy},$ E is the nature radix, e=2.71828 ..., y is variable.

Images