JP5213997B2 - Method and apparatus for coexistence detection - Google Patents

Description

  The present invention relates to coexistence detection in a communication system, and more particularly to coexistence detection and signal detection in IP routing, for example.

60 GHz communication is a new short range wireless communication technique that supports gigabit transmission with a relatively wide operating bandwidth (4-7 GHz).
Currently, many standards for 60 GHz millimeter wave WLAN / WPAN (Wireless Local Area Network / Wireless Personal Area Network) have already been established for various applications (eg IEEE 802.15.3c, WirelessHD, WiGig, etc.). Yes. Therefore, coexistence detection of various 60 GHz systems that occupy the same frequency band is important.

The coexistence detection methods currently used in 60 GHz systems can be divided into two categories.
One is an ED (Energy Detection) method (H. Urkowits, “Energy detection of unknown deterministic signals” in Processing of the IEEE, vol. 55, Apr. 1967, pp. 523-531 (Non-patent Document 1). And the other is a CMS (Common Mode Signaling) system defined in IEEE 802.15.3c.

H. Urkowits, "Energy detection of unknown deterministic signals" inProcessing of the IEEE, vol. 55, Apr. 1967, pp. 523-531 M. Park, E. Perahia and M. Gong, “Analysis on IEEE 802.15.3ccoexistence scheme” in IEEE 802.11-09 / 559r0, May 2009 S. Kato, CS Sun, T. Baykas, Z. Lan, J. Wang, R. Funada, MA Rahman, CW Pyo, H. Harada and I. Lakkis, “Common mode signaling (CMS) for inter-system coexistence enhancement” in IEEE 802.11-09-370-00-00ad, May 2009 Y. Zeng and L. C. Liang "Eigenvalue-based spectrum sensing algorithm for cognitive radio" in IEEE Transactions on Communications, vol.57, Jun.2009, pp.1784-1793 R. Zhang, T. J. Lim, Y. C. Liang, and Y. Zeng “Multi-antenna based spectrum sensing for cognitive radios: a GLRT approach” in IEEE Transactions on Communications, vol.58, Jan. 2010, pp. 84-88 W. S. Liggett, “Passive sonar: fitting models to multiple time series” in Processing of the NATO ASI on Signal Processing, J. W. R Griffiths and P. L. Stocklin, Eds. Academic Press: New York, 1973, pp.327-345

  The above-mentioned ED is the most commonly used signal detection method due to its simplicity and effectiveness. However, conventional EDs are affected by noise uncertainty and have insufficient detection performance in a low SNR (signal to noise ratio) state. Such a phenomenon is more prominent under the 60 GHz channel environment (M. Park, E. Perahia and M. Gong, “Analysison IEEE 802.15.3c coexistence scheme” in IEEE 802.11-09 / 559r0, May 2009 (non-patent Reference 2)).

  The CMS described above involves complex PHY / MAC level adjustment of the control frame, which makes it very expensive and difficult to implement in practical applications (S. Kato, CS Sun, T. Baykas, Z. Lan, J. Wang, R. Funada, MA Rahman, CW Pyo, H. Harada and I. Lakkis, “Commonmode signaling (CMS) for inter-system coexistence enhancement” in IEEE 802.11-09-370-00-00ad , May 2009 (see Non-Patent Document 3)).

  Therefore, there is a need for a coexistence detection method that is superior in performance to the ED used in the current 60 GHz system and that reduces the complexity of PHY / MAC level adjustment of control frames as compared to CMS.

  The present invention provides a new two-stage coexistence detection method for 60 GHz MMW (millimeter wave) WLAN / WPAN. The energy detector is used in the first stage of performing a rough scan. If it is detected that there is a signal transmitted from a communication device that uses another communication standard that occupies the channel, transmission of a device that needs to transmit payload data is avoided. Otherwise, the second stage detector is activated. If it is detected in the second stage that the channel is empty, devices that need to transmit payload data are allowed to transmit. Otherwise, transmission of payload data is avoided.

  According to the present invention, a method for performing coexistence detection on a communication system that occupies the same channel for data transmission includes calculating a first detection statistic according to a received signal, and wherein the first detection statistic is Calculating a second detection statistic if it is not greater than a predetermined threshold of 1, and if the second detection statistic is not greater than a second predetermined threshold And determining that one of the first and second detection statistics values is calculated based on received signal energy and noise variance; and The other of the first and second detection statistics is calculated based on a plurality of eigenvalues of the sample covariance matrix of the received signal.

  According to the coexistence detection method of the present invention, the first detection statistical value is larger than the first predetermined threshold value, or the second detection statistical value is larger than the second predetermined threshold value. In this case, it is determined that the channel is occupied and a notification is issued to avoid transmission of payload data.

  Preferably, the first predetermined threshold is set based on a known false alarm probability and a distribution of the first detection statistics value, and the second predetermined threshold is a known false alarm probability. And based on the distribution of the second detection statistical value.

  Preferably, the other of the first and second predetermined threshold values is a logarithmic function of a ratio of an arithmetic average to a geometric average of a plurality of eigenvalues.

  Preferably, the distribution of the first and second detection statistics values follows a chi-square distribution with different degrees of freedom.

  An apparatus for performing coexistence detection on a communication system occupying the same channel for data transmission according to the present invention comprises a first detector and a second detector, wherein the first detector is a received signal. And a first calculating means for calculating a first detection statistical value according to the above, and calculating a second detection statistical value when the first detection statistical value is not greater than a first predetermined threshold value. First comparison determination means, wherein the second detector calculates a second detection statistical value according to the received signal, and the second detection statistical value is a second predetermined value. A second comparison / determination unit for determining that the channel is not occupied and notifying permission of transmission of payload data when not greater than the threshold, and one of the first and second detection statistics Is calculated based on the received signal energy and noise variance, 1 and the other of the second detection statistic is calculated based on the plurality of eigenvalues of the sample covariance matrix of the received signal.

  Preferably, the second detection statistic value is greater than the first predetermined threshold value or the second detection statistic value is greater than the second predetermined threshold value. The comparison determination means determines that the channel is occupied and issues a notification that avoids transmission of payload data.

  Preferably, the first predetermined threshold is set based on a known false alarm probability and a distribution of the first detection statistics value, and the second predetermined threshold is a known false alarm probability. And based on the distribution of the second detection statistical value.

  Preferably, the other of the first and second predetermined threshold values is a logarithmic function of a ratio of an arithmetic average to a geometric average of a plurality of eigenvalues.

  Preferably, the distribution of the first and second detection statistics values follows a chi-square distribution with different degrees of freedom.

  The two-stage coexistence detection method proposed in the present invention has better detection performance than the ED currently used in the 60 GHz channel. Also from the viewpoint of computational complexity, the two-stage coexistence detection method of the present invention significantly reduces the complexity of PHY / MAC level adjustment of control frames compared to CMS.

The above and other features and advantages of the present invention will become more apparent from the following preferred embodiments described with reference to the drawings.
It is a figure which shows the interference model of 60 GHz WLAN / WPAN. It is a figure which shows schematic structure of the step coexistence detection apparatus by preferable embodiment of this invention. 3 is a flowchart illustrating a two-stage coexistence detection method according to a preferred embodiment of the present invention.

Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. In the embodiment, a 60 GHz WLAN / WPAN system will be described as an example.
However, it will be appreciated by those skilled in the art that the present invention is not limited to the examples shown in the embodiments. The embodiments are merely examples and are not limited thereto. The scheme of the present invention can be applied to any communication system operating on the same frequency, and is within the protection scope of the present invention.

FIG. 1 shows a 60 GHz WLAN / WPAN 100, which is an exemplary communication system to which the present invention is applied. In the 60 GHz WLAN / WPAN 100, two WLANs are shown as examples. Each WLAN includes one WLAN controller and a number of 60 GHz devices. These two WLANs operate in the same frequency band using different 60 GHz standards (for example, WiGig (Wireless Gigabit) for WLAN1 and IEEE 802.15.3c for WLAN2). The devices 2A and 2B in the WLAN 2 may interfere with the devices 1C and 1D in the WLAN 1.
FIG. 1 shows the interference area.
Before it becomes necessary to establish a communication link by a device in WLAN1, the WLAN1 controller first scans the entire spectrum. Since the devices in WLAN1 and WLAN2 operate in the same frequency band, if the WLAN1 controller finds out that the transmission channel is occupied by the coexistence detection, transmission by the device necessary for transmitting payload data is avoided. Alternatively, the transmission is retuned to minimize intersystem interference.

  Hereinafter, the configuration of the two-stage coexistence detection apparatus according to a preferred embodiment of the present invention will be described with reference to FIG.

FIG. 2 shows a two-stage coexistence detection apparatus 200 according to a preferred embodiment of the present invention.
The two-stage coexistence detection device 200 includes an energy detector 201 and a full blind detector 203. The energy detector 201 includes a first calculation unit 2011 and a first comparison determination unit 2013. The full blind detector 203 includes second calculation means 2031 and second comparison determination means 2033. It should be noted that in FIG. 2, only the components related to the present invention in the two-stage coexistence detection device 200 are shown for easy understanding, and the well-known components are omitted.

The two-stage coexistence detection apparatus 200 according to the present invention is compatible in a 60 GHz system regardless of a single carrier (SC) 60 GHz system or a multicarrier (eg, OFDM) 60 GHz system. Used for detection.
In the following, the two-stage coexistence detection process according to the present invention will be described with reference to FIG. 3, taking as an example a case where it is applied to an OFDM 60 GHz system.

FIG. 3 shows a flowchart of a two-stage coexistence detection method 300 according to a preferred embodiment of the present invention.
Here, two assumptions H ₀ and H ₁ in statistics are applied. Here, H ₀ represents an assumption that there is no signal that occupies a channel transmitted from a communication device that uses another communication standard, and H ₁ that there is a signal that occupies a channel. It represents an assumption.

First, in step S301, the first calculation means 2011 in the energy detector 201 in the first stage calculates a corresponding energy detection statistical value T _ED for the received signal y.

In the exemplary OFDM 60 GHz system, the received signal y is a vector.
Assuming y _n (n = 0,…, N-1) is observed in N independent observation blocks during the observation period, and each y _n is an Mx1 vector, they are statistically independent of each other. Yes.
_Let M _n measurements in y _{n be} y _n = [y _n (0), ..., y _n (M-1)] ^T , and
H ₀ : y _n = w _n , n = 0,…, N-1
H ₁ : y _n = x _n + w _n ,
n = 0, ..., N-1.
And

That is, under assumption H ₀ , the received signal y _n is only a complex Gaussian vector in the system.
Here, each component follows a complex Gaussian distribution CN (0, σ ² ) having an average of 0 and a variance σ ² . Meanwhile, under the assumption H1, the received signal y _n is the sum of the transmission signal x _n to be detected as complex Gaussian noise vector in the system. Here, x _n is a complex Gaussian distribution with an average of 0, and the dispersion matrix is assumed to be expressed as R _x = E [x _n x _n ^H ].

Specifically, in step S301, the energy detection statistical value T _ED is calculated by the first calculation unit 2011 in the energy detector 201 according to the following equation.

In step S303, the first comparison / determination means 2013 in the energy detector 201 compares the calculated energy detection statistical value T _ED with a predetermined threshold τ in energy detection.

The predetermined threshold τ is defined during system design in relation to the distribution of acceptable probability of false alarm P _F ′ and energy detection statistics T _ED .
The decision threshold value (that is, the predetermined threshold value) τ in energy detection is set according to the following method.

である。
ここで、Γ(・,・)は、不完全ガンマ関数を表し、Γ(・)はガンマ関数を表す。
従って、誤報確率P_F'は、以下の式に従って計算される。
P_F'=Pr{T_ED>τ|H₀}=1-F_TED|H0(τ) Under H ₀ , T _ED follows a chi-square distribution with 2 MN degrees of freedom.
Therefore, under H ₀ , the CDF (Cumulative Density Function) of T _ED is

It is.
Here, Γ (·, ·) represents an incomplete gamma function, and Γ (·) represents a gamma function.
Accordingly, the false alarm probability P _F ′ is calculated according to the following formula.
P _F '= Pr {T _ED > τ | H ₀ } = 1-F _{TED | H0} (τ)

So, for a given P _F ′, the decision threshold τ in energy detection is
τ = F ^-1 _{TED | H0} (1-P _F ') (2)
And get.
Here, F ^-1 _{TED | H0} (•) is an inverse function of F _{TED | H0} (•).

In step S303, if the first comparison / determination means 2013 determines that T _ED is larger than τ, it corresponds to H ₁ , thereby enabling a communication device using another communication standard to occupy the channel. It is determined that there is a transmitted signal.
In step S313, the first comparison determination unit 2013 notifies that transmission of a device that needs to transmit payload data is avoided.

In step S303, if the first comparison / determination means 2013 determines that T _ED is not larger than τ, it corresponds to H ₀ , thereby transmitting from a communication device using another communication standard that occupies the channel. It is determined that the processed signal does not exist.
Next, in step S <b> 305, the first comparison determination unit 2013 transmits an enable signal to the full blind detector 203 in order to operate the full blind detector 203.

  In the present invention, an AGD (arithmetic mean to geometric mean detector) is arranged in the second stage in order to realize full blind detection (Y. Zeng and LC Liang). "Eigenvalue-based spectrum sensing algorithm for cognitive radio" in IEEE Transactions on Communications, vol.57, Jun.2009, pp.1784-1793 (Non-Patent Document 4)).

The full blind detector AGD 203 installed in the second stage strictly follows the GLRT (Generalized Likelihood Ratio Test) paradigm to derive the detection statistic T _AGD .
AGD that does not require any prior knowledge of R _x and σ ² is completely blind (R. Zhang, TJ Lim, YC Liang, and Y. Zeng “Multi-antenna based spectrum sensing for cognitive radios: a GLRT approach ”in IEEE Transactions on Communications, vol.58, Jan. 2010, pp. 84-88 (Non-Patent Document 5)).
The AGD checks the inherent relevance of the received data samples (particularly the eigenspace of the received data sample variance matrix). Therefore, AGD has robustness against noise uncertainty.

As shown in FIG. 3, in step S307, the second calculation means 2031 in the full-blind detector AGD 203 in the second stage responds to the received signal y based on the formula derived according to the GLRT paradigm. Calculate the full blind detection statistic T _AGD .
In a typical OFDM 60 GHz system, the received signal y must be a vector.
Similar to the assumption in energy detection performed in the first stage, y _n (n = 0,..., N−1) is observed in N independent observation blocks during the observation period, and each y _n is an Mx1 vector. Are statistically independent of each other.
_Let M _n measurements in y _{n be} y _n = [y _n (0), ..., y _n (M-1)] ^T , and
H ₀ : y _n = w _n , n = 0,…, N-1
H ₁ : y _n = x _n + w _n ,
n = 0, ..., N-1.
And

ここで、Ｋは、乗算因子（H₀の下でのフルブラインド検出統計値T_AGD
の分布を修正するために用いられる）であり、λ_m={λ₀,…,_λM-1}は、降順におけるy_nの標本共分散行列

の固有値分解によって取得されたＭ個の固有値である。
また、

は、それぞれ固有値λ_mのＡＭ（算術平均）とＧＭ（幾何平均）である。 Specifically, in step S301, the detection statistical value T _ED is calculated by the second calculation means 2031 in the full blind detector 203 based on the following formula derived according to GLRT.

Where K is a multiplication factor (full blind detection statistic T _AGD under H ₀
Λ _m = {λ ₀ , ..., _λM-1 } is the sample covariance matrix of y _n in descending order

M eigenvalues obtained by eigenvalue decomposition of.
Also,

Are AM (arithmetic mean) and GM (geometric mean) of eigenvalues λ _m , respectively.

Next, in step S309, the second comparison determination unit 2033 in the energy detector 203 compares the calculated detection statistical value T _AGD with a predetermined threshold value η in full blind detection.

The predetermined threshold value η is also related to the distribution of _receivable false alarm probability P _F ″ and detection statistical value T _AGD .
The decision threshold (that is, the predetermined threshold) η in the full blind detection is set by the following method.

と設定される（W. S. Liggett, “Passive sonar: fitting models to multiple time series” in Processing of the NATO ASI on Signal Processing, J. W. R Griffiths and P. L. Stocklin, Eds. Academic Press: New York, 1973, pp.327-345（非特許文献６）を参照）。
これにより、H₀の下におけるT_AGDのCDFは、次のように表わされる。

ここで、Γ(・,・)は、不完全ガンマ関数を表し、Γ(・)はガンマ関数を表す。
従って、フルブラインド検出における誤報確率P_F''は、P_F''=1-F_TAGD|H0(η)として計算される。 As described above, the multiplication factor K is used to modify the distribution of the full blind detection statistic T _AGD under H ₀ .
Where K is asymptotically followed by a chi-square distribution with T _AGD (M ² -1)

(WS Liggett, “Passive sonar: fitting models to multiple time series” in Processing of the NATO ASI on Signal Processing, JW R Griffiths and PL Stocklin, Eds. Academic Press: New York, 1973, pp.327- 345 (see Non-Patent Document 6).
Thus, the CDF of T _AGD under H ₀ is expressed as follows:

Here, Γ (·, ·) represents an incomplete gamma function, and Γ (·) represents a gamma function.
Therefore, the false alarm probability P _F ″ in full blind detection is calculated as P _F ″ = 1−F _{TAGD | H0} (η).

Therefore, for a given P _F '', the decision threshold η in full blind detection is
η = F _{TAGD | H0} ^-1 (1- P _F '') (4)
Get as.
here,
F _{TAGD | H0} ⁻¹ (•) is an inverse function of F _{TAGD | H0} (•).

In step S309, if the second comparison and determination unit 2033 determines that T _AGD is greater than η, it corresponds to H ₁ , whereby a signal transmitted from a communication device using another communication standard that _occupies the channel is received. It is determined that it exists.
In step S313, the second comparison determination unit 2033 notifies that transmission of a device that needs to transmit payload data is avoided.

If the second comparison / determination means 2033 determines in step S309 that T _AGD is not larger than η, it corresponds to H ₀ , and thereby a signal transmitted from a communication apparatus using another communication standard that _occupies the channel. Is determined not to exist.
Next, in step S311, the second comparison determination unit 2033 notifies that transmission of a device that needs to transmit payload data is executed.

Thus, according to the two-stage coexistence detection method of the present invention, if the energy detector 201 mistakenly declares that the channel is not occupied in the first stage, the full-blind detector 203 is erroneous in the second stage. It is possible to correct the decision.
Activating the full-blind detector 203 only when the energy detector 201 informs that the channel is not occupied in the first stage, compared to the method in which the full-blind detector 203 is directly used for coexistence detection Therefore, according to the two-stage coexistence detection method of the present invention, the computational complexity is reduced.

  Although the above embodiment has been described for application in a multi-carrier (OFDM) 60 GHz system, it will be appreciated by those skilled in the art that it can be similarly applied in a single-carrier (SC) 60 GHz system.

In addition, the above embodiment of the present invention has explained that the coexistence detection is realized by the energy detector in the first stage and the full blind detector in the second stage. Is not limited to the special constraints described above.
It will be appreciated by those skilled in the art that coexistence detection can also be achieved with a full blind detector in the first stage and an energy detector in the second stage without taking into account the computational complexity. It will be.
In such an embodiment, if the full blind detector used in the first stage detects the presence of a signal transmitted from a communication device that uses another communication standard that occupies the channel, Transmission of devices that need to transmit data is avoided. Otherwise, the energy detector in the second stage is activated.
In the second stage, if it is detected that the channel is empty, devices that need to transmit payload data are allowed to transmit. Otherwise, transmission of payload data is avoided.

Those skilled in the art will readily understand that the steps and operations in the above method can be implemented by a software program.
Such implementations further include a machine readable or computer readable program storage device (such as a digital data storage medium, etc.) and program instructions executable by a programming machine or computer, the instructions being as described above. Perform some or all of the steps in the method.
For example, the program storage device is a digital storage device, a magnetic storage medium such as a magnetic tape or a magnetic disk, hardware, or an optically readable digital data storage medium. This implementation further includes a programming computer for performing the steps in the method.

  Although the present invention has been described with reference to preferred embodiments thereof, various other modifications, changes and additions can be made by those skilled in the art without departing from the spirit and scope of the present invention. It will be clear that this is possible. Accordingly, the scope of the invention is not limited to the specific embodiments described above, but only by the appended claims.

200: Two-stage coexistence detection apparatus 201: Energy detector 2011: First calculation means 2013: First comparison determination means 203: Full blind detector 2031: Second calculation means 2033: Second comparison determination means

Claims (10)

A method for performing coexistence detection on a communication system occupying the same channel for data transmission, comprising:
Calculating a first detection statistic according to the received signal;
Causing the second detection statistic to be calculated if the first detection statistic is not greater than a first predetermined threshold;
Determining that the channel is not occupied if the second detection statistic is not greater than a second predetermined threshold, and notifying permission to transmit payload data;
One of the first and second detection statistics is calculated based on received signal energy and noise variance;
The other of the first and second detection statistics is calculated based on a plurality of eigenvalues of the sample covariance matrix of the received signal.   A channel is occupied when the first detection statistic is greater than the first predetermined threshold or the second detection statistic is greater than the second predetermined threshold. The coexistence detection method according to claim 1, wherein a notification for determining and avoiding transmission of payload data is issued. The first predetermined threshold is set based on a known false alarm probability and a distribution of the first detection statistics,
The coexistence detection method according to claim 1, wherein the second predetermined threshold is set based on a known false alarm probability and a distribution of the second detection statistical value.   The coexistence detection method according to claim 1, wherein the other of the first and second predetermined threshold values is a logarithmic function of a ratio of an arithmetic average to a geometric average of a plurality of eigenvalues.   The coexistence detection method according to claim 1, wherein the distribution of the first and second detection statistics values follows a chi-square distribution with different degrees of freedom. An apparatus for performing coexistence detection on a communication system that occupies the same channel for data transmission,
Comprising a first detector and a second detector;
The first detector comprises:
First calculating means for calculating a first detection statistic according to the received signal;
A first comparison / determination means for calculating a second detection statistic when the first detection statistic is not greater than a first predetermined threshold;
The second detector comprises:
Second calculating means for calculating a second detection statistic according to the received signal;
A second comparison / determination unit that determines that the channel is not occupied when the second detection statistic value is not greater than a second predetermined threshold and notifies the transmission permission of payload data; ,
One of the first and second detection statistics is calculated based on received signal energy and noise variance;
The other of the first and second detection statistics is calculated based on a plurality of eigenvalues of the sample covariance matrix of the received signal.   The second comparison / determination means when the first detection statistical value is larger than the first predetermined threshold value or the second detection statistical value is larger than the second predetermined threshold value. The coexistence detection apparatus according to claim 6, wherein the coexistence detection apparatus determines that the channel is occupied and issues a notification for avoiding transmission of payload data. The first predetermined threshold is set based on a known false alarm probability and a distribution of the first detection statistics,
The coexistence detection apparatus according to claim 6, wherein the second predetermined threshold is set based on a known false alarm probability and a distribution of the second detection statistical value.   7. The coexistence detection apparatus according to claim 6, wherein the other of the first and second predetermined threshold values is a logarithmic function of a ratio of an arithmetic average to a geometric average of a plurality of eigenvalues. The coexistence detection apparatus according to claim 6, wherein the distribution of the first and second detection statistics values follows a chi-square distribution with different degrees of freedom.

Images