Connect public, paid and private patent data with Google Patents Public Datasets

Cognitive radio master user location method based on backtracking check convex set projection

Download PDF

Info

Publication number
CN103200670A
CN103200670A CN 201310057842 CN201310057842A CN103200670A CN 103200670 A CN103200670 A CN 103200670A CN 201310057842 CN201310057842 CN 201310057842 CN 201310057842 A CN201310057842 A CN 201310057842A CN 103200670 A CN103200670 A CN 103200670A
Authority
CN
Grant status
Application
Patent type
Prior art keywords
projection
user
iteration
convex
set
Prior art date
Application number
CN 201310057842
Other languages
Chinese (zh)
Other versions
CN103200670B (en )
Inventor
杜利平
雷雨
康璐璐
姜少坤
Original Assignee
北京科技大学
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date

Links

Abstract

The invention provides a cognitive radio master user location method based on backtracking check convex set projection. L coordinates for sensing a user are regarded as a circle center, a convex set projection algorithm is used for carrying out Mc step rectangular projection iteration on the a master user; mc step backtracking comparing checking is carried out, the distances between adjacent iteration points are computed and are compared with a threshold value lambada; if parts of the distances between the adjacent iteration points are zero or larger than lambada, Mb step projection iteration and mb step backtracking comparing checking on the border of a convex set circular region are carried out, distance values between two adjacent iteration points are computed and are compared with the threshold value lambada again, and if the distance values are all smaller than lambada, the Mb step iteration result bMb is determined as a locating result of position information of the master user. According to the backtracking check convex set projection location algorithm, shortcomings of an existing convex set projection location algorithm are overcome, the location algorithm is good, influence from distance-measuring errors is small, and the method is suitable for a link in which a sensing user acquires the location information of the master user in a cognitive radio network.

Description

基于回溯检查凸集投影的认知无线电主用户定位方法 Check back POCS based cognitive radio primary user positioning method

技术领域 FIELD

[0001] 本发明涉及一种认知无线电网络中对主用户定位的方法,特别涉及一种基于回溯检查的凸集投影定位方法。 [0001] The present invention relates to a method for cognitive radio networks primary user location, particularly to a location based on projections onto convex sets back check.

背景技术 Background technique

[0002] 随着无线移动通信与计算机网络的结合应用发展越来越成熟,移动互联网已经称为当今世界发展速度最快、市场潜力最大,商业价值最高的发展业务之一。 [0002] With the development of wireless applications in conjunction with mobile communications and computer networks more and more sophisticated, mobile Internet has been called the world's fastest growing and one of the highest development of the business market with the greatest potential commercial value. 丰富的应用主要依托于文字、图像、视频等信息承载方式,而且随着人们对应用质量的不断追求,要求信息的传递越来越高效、便捷。 Rich applications mainly relying on the information bearer text, images, video, etc., but as people continue to pursue the application of quality, requiring the transmission of information more efficient and convenient. 这些应用的创新与发展需要较宽频谱和较高下载速率。 Innovation and development of these applications require a wide spectrum and higher download speeds. 认知无线电的概念迎合了频谱重复利用的需要,可以通过对无线环境的感知实现与主用户的冲突避免,利用最优化的决策有效的动态利用频谱空洞。 The concept of cognitive radio spectrum reuse to meet the need, we can avoid the use of optimized decision-making effective use of spectrum holes by dynamic perception of the wireless environment to achieve the primary user of the conflict. 如果能够获取主用户的位置信息,那么对频谱感知的性能将会得到很大提升,并且在后续的频谱资源的管理与分配中也将起到很大的帮助作用。 If you can get the location information of the master user, then the spectrum sensing performance will be greatly improved, and will also play a very helpful role in the follow-up of spectrum management and distribution resources.

[0003] 在认知网络中对主用户进行定位,获取主用户的位置信息的主要作用有以下几个方面: The main role of [0003] the user to position the primary cognitive network obtains location information of the user has the following main aspects:

[0004] 1.为频谱资源管理提供支持。 [0004] 1. To provide support for the management of spectrum resources. 在主用户位置信息已知的情况下,根据其位置信息可以更好地提高频谱利用率,更好地指导感知用户不干扰主用户的频谱使用。 In the main user position information it is known, according to which the location information to better spectrum efficiency, better guide the user does not perceive the interference spectrum using the primary user.

[0005] 2.减小认知网络中用户的功耗。 Power [0005] 2. The decreases in cognitive network users. 在主用户位置信息已知的情况下,认知网络中的感知用户可以根据主用户的位置信息来确定频谱感知的方向,在最小功率的运行状态下,便可以准确判断主用户的频谱使用情况。 In the main user position information is known, the user perception of the cognitive network may be determined based on the spectrum sensing direction of the position information of the primary user, in the operating state of minimal power, it can accurately determine the spectrum using the primary user's situation .

[0006] 3.避免对主用户的干扰。 [0006] 3. To avoid interference to the primary user. 在主用户位置信息已知的情况下,可以结合多天线技术,针对主用户的方向位置进行频谱感知,避免了频谱间相互干扰的可能性。 In the main user position information is known, it may be combined with multi-antenna technology, spectrum sensing for the position of the main direction of the user, and avoids the possibility of inter-spectrum interference.

[0007] 4.有利于感知用户的位置优化。 [0007] 4. facilitate user optimization of perceptual location. 在主用户位置信息已知的情况下,根据主用户的位置信息,可以合理的分布感知用户的位置,提高频谱和空间的利用率,更好地避免对主用户干扰。 In the main user position information is known, the position information of the primary user, the user can reasonably distributed sensing position, and improving the spectrum utilization of space, better avoid interference to the primary user.

[0008]目前常用的凸集投影方法包括 Circular POCS,Hyperbolic POCS,Boundary POCS和Hybrid POCS等,其中Hybrid POCS是前两种POCS方法的合并,根据研究结果表明,Hybrid POCS方法的定位精度要优于前几种方法,然而,当主用户远离感知用户时,由于Hybrid POCS算法中双曲线投影定位对于主用户在感知用户多边形之外的情况下收敛点受噪声波动较大,因此误差随着测距增大而增大。 [0008] The most commonly used method comprises POCS Circular POCS, Hyperbolic POCS, Boundary POCS POCS the Hybrid and the like, where the merger of the first two is the Hybrid POCS POCS method, according to the research results show that the positioning accuracy POCS method is superior to the Hybrid several prior methods, however, when the main user away from the user perception, since the positioning projection hyperbolic Hybrid POCS algorithm for the primary user other than the user perception of the polygon convergence point by noise fluctuations, so the error increases as the distance large increases.

发明内容 SUMMARY

[0009] 本发明旨在解决上述技术缺陷,提出一种应用于认知网络中对主用户进行定位的回溯凸集投影算法(BackCheck P0CS)。 [0009] The present invention is intended to solve the above technical defects, it is applied to provide a cognitive network backtracking algorithm for projections onto convex sets (BackCheck P0CS) positioning the main user.

[0010] 该方法包括以下步骤: [0010] The method comprises the steps of:

[0011] 步骤一、以L个感知用户的坐标为圆心,利用凸集投影算法对主用户进行Mc步正交投影迭代,得到Mc个迭代点X k ,其中k = 1,2,3,..., Mc ; [0011] Step a, to the user's perception of L as the center coordinates, using the POCS algorithm primary user Mc iteration step the orthogonal projection, to give Mc iteration points X k, where k = 1,2,3 ,. .., Mc;

[0012] 步骤二.对步骤一中获取的Mc个迭代点,进行me步回溯比较检查,计算出相邻迭代点之间的距离II xm+1-xm II,其中,m=Mc-l,..., Mc-mc [0012] Step II. Mc iteration of points acquired in a step, the step for me back comparison check, calculate the distance between adjacent iterates II xm + 1-xm II, where, m = Mc-l, ..., Mc-mc

[0013] 步骤三.如果步骤二中的回溯比较检查中,相邻迭代点之间的距离均小于λ且不为零,则将步骤一中最后L个迭代均值作为主用户位置信息的定位结果;如果步骤二中的回溯比较检查中,相邻迭代点之间的距离存在部分为零或大于λ的情况,继续执行步骤四; [0013] Step III. If the comparison check back in step two, the distance between adjacent points are less than λ iterative and non-zero, then a last step of L iterations average primary user position information as the positioning result ; step II if the backtracking comparison check, zero or greater than the partial distance λ between adjacent iterations points exist, proceed to step four;

[0014] 步骤四.以Mc步迭代结果χΜ。 [0014] Step Four. ΧΜ to Mc iteration results. 为初始点Iv进行凸集圆域边界上正交投影迭代,迭代检查步数为Mb,得至IJ Mb个迭代点bh,其中h=l, 2,3,...Mb ; Iv is the initial point for the iterative orthogonal projection on convex sets gamut boundary circle, checking the iteration step number Mb, to obtain IJ Mb iterations points BH, where h = l, 2,3, ... Mb;

[0015] 步骤五.对步骤四中获取的Mb个迭代点,进行mb步回溯比较检查,计算相邻两个迭代点间的距离值I |bn+1-bn| I,其中,n=Mb-l,...,Mb-mb,并与门限值λ进行比较; . [0015] Step five pairs Mb iteration step four points acquired, for comparison check mb step back, I calculated distance value between adjacent two iterates | bn + 1-bn | I, where, n = Mb -l, ..., Mb-mb, and compared with a threshold value [lambda];

[0016] 步骤六.如果步骤五中的回溯比较检查中,相邻迭代点间的距离值均小于λ,则将步骤四中最后L个迭代均值作为主用户位置信息的定位结果;如果步骤五中的回溯比较检查中,相邻迭代点间的距离值存在大于λ的情况,跳转到步骤四以Mb步迭代结果bi为初始点K,并变换投影迭代顺序,直到相邻迭代点间的距离值均小于λ。 [0016] Step Six If step back comparison inspection Fifth, the distance between adjacent iterates the value is less than [lambda], then the last L four-step iteration of the mean result of the primary user as a positioning position information; If step five the comparison check back, the distance value exists between adjacent points is larger than λ iteration, the four jumps to step to result bi K Mb iteration as the initial point, and converting projection iteration order, between adjacent points until the iteration distance values ​​are less than λ.

[0017] 优选地,所述步骤一包括: [0017] Preferably, said step a comprises:

[0018] 1.1)初始化步骤:设置初始点X(i,其中XtI为任意位置上的一点; [0018] 1.1) initialization step: setting the initial point X (i, XtI to a point where an arbitrary position;

[0019] 1.2)利用以下公式进行投影迭代: [0019] 1.2) using the following equation iteration are projected:

[0020] [0020]

Figure CN103200670AD00051

[0022] 其中表示正交凸集投影点,di[cos(φ),sin (φ)]T表示Pi到Pi+1的向量A为第i个感知用户的位置坐标,ie [l,L]。 [0022] wherein represents the point of orthogonal projection onto convex sets, di [cos (φ), sin (φ)] T represents the vector Pi A Pi + 1 is the i-th user's perceived position coordinate, ie [l, L] . Di为以第i个感知用户为圆心,以第i个感知用户测得的与主用户之间的距离测量值为半径的凸集圆域。 Di is the i-th to user perceived as the center, the distance between the measured value to the radius of an i-th user perceived measured round convex set primary user domain.

[0023] 优选地,λ的取值大小取决于感知用户对主用户测距结果的平均值,且λ相对于该平均距离是一个很小值。 [0023] Preferably, the value of [lambda] depends on the user perception of the average distance to the primary results of the user, and with respect to the average distance [lambda] is a very small value.

[0024] 优选地,mb与me取值相同,为L的整数倍。 [0024] Preferably, the mb me same value as an integer multiple of L.

[0025] 优选地,所述步骤四包括: [0025] Preferably, said four comprising the step of:

[0026] 4.1)初始化步骤:I)设置初始点b。 [0026] 4.1) initialization step: I) the initial set point b. ,b。 , B. = xMc = XMc

[0027] 4.2)利用以下公式进行投影迭代: [0027] 4.2) using the following equation iteration are projected:

Figure CN103200670AD00052

[0030] 其中,Pi为第i个感知用户的位置坐标,ie [I, L]。 [0030] wherein, Pi is the i-th user's perceived position coordinate, ie [I, L]. Cli是第i个感知用户测得的与主用户之间的距离测量值;Ci={ye R2: I y-Pj I =ClJ是第i个感知用户所确定的半径为(Ii的圆边界。 Cli distance measurements between the user perception of the i-th measured primary user; Ci = {ye R2: I y-Pj I = ClJ i th user perception of the determined radius (boundary circle of Ii.

[0031] 该算法基于凸集投影定位算法的改进,弥补了现有凸集投影定位算法的不足,受测距误差的影响相对较小,适合应用于认知无线电网络中感知用户对主用户位置信息的获取环节,能够更准确的实现对主用户的定位。 [0031] The algorithm is based Convex Sets improved positioning projection algorithm, to make up for deficiencies in the prior positioning POCS algorithm, affected by the distance error is relatively small, suitable for cognitive radio networks perceived by the user to the primary subscriber location obtaining link information, we can achieve more accurate positioning of the primary user.

附图说明 BRIEF DESCRIPTION

[0032] 图1为本发明中基于BackCheck POCS定位方法的流程图。 [0032] FIG. 1 is a flowchart of the positioning method based BackCheck POCS invention.

[0033] 图2为本发明基于BackCheck POCS定位方法的迭代不意图。 [0033] FIG. 2 BackCheck POCS iteration based positioning method of the present invention is not intended.

[0034] 图3为BackCheck POCS和Hybrid POCS的定位误差对比示意图。 [0034] FIG. 3 is a positioning error and BackCheck POCS Hybrid POCS schematic comparison.

[0035] 图4为不同测距误差对BackCheck POCS及Hybrid POCS定位误差影响的对比示意图。 [0035] 4 is a schematic comparison of different BackCheck POCS ranging error and positioning error on Hybrid POCS FIG.

[0036] 下面通过参考附图描述的实施例是示例性的,仅用于解释本发明,而不能解释为对本发明的限制。 [0036] Next, by way of example with reference to the accompanying drawings are exemplary only for explaining the present invention and should not be construed as limiting the present invention.

[0037] 假设有L个感知用户参与对主用户的定位,L个感知用户的位置坐标是已知的,表示为: [0037] Suppose there are L user perceived involvement of positioning of the primary user, the position coordinates of the user's perception of the L are known, is expressed as:

Figure CN103200670AD00061

[0039] L个感知用户测得的与主用户之间的距离测量值表示为: [0039] The distance L between the measured value measured perceptual user and a primary user is expressed as:

Figure CN103200670AD00062

[0041] 以每个感知用户为圆心,以距离测量值Cli为半径的凸集圆域表示为: [0041] In each of the user perceived as the center, distance measurement Cli circle whose radius is convex set domain representation of:

Figure CN103200670AD00063

[0043] 第i个感知用户所确定的半径为Cli的圆边界表示如下: [0043] i-th user perception of the determined radius of the boundary circle Cli expressed as follows:

Figure CN103200670AD00064

[0045] 本发明定位方法所用的POCS算法可以是Circular POCS, Hyperbolic POCS和Boundary POCS。 [0045] The method of the present invention positioned POCS method may be used Circular POCS, Hyperbolic POCS and Boundary POCS. 现以Circular POCS算法为例,给出基于BackCheck POCS算法的具体步骤: Circular POCS algorithm is to an example is given based on the specific steps BackCheck POCS algorithm:

[0046] 步骤一.以L个感知用户的坐标为圆心,利用Circular POCS对主用户进行Mc步正交投影迭代,得到Mc个迭代点X1' x2、...xMc_!> xMc。 [0046] Step a. In the L perceived as the center coordinates of the user, the main user using iterative orthogonal projection Mc step Circular POCS, Mc obtained iteration points X1 'x2, ... xMc _!> XMc.

[0047] 在此设定Circular POCS的迭代检查步数为Mc,由于Circular POCS的收敛速度很快,Mc的取值可以适当取一个较小值。 [0047] In this setting number of checks Circular POCS iteration of Mc, since the convergence rate Circular POCS quickly, Mc values ​​may suitably take a smaller value.

[0048] 根据Circular POCS的投影迭代规则,对主用户进行Mc步正交投影迭代的步骤为: [0048] The projection rules Circular POCS iteration, the primary user Mc step iteration is the orthogonal projection steps:

[0049] I)初始化:设置初始点Xtl,其中Xtl为平面内任意一点,如图2中方块所示; [0049] I) initialization: setting the initial point Xtl, wherein Xtl any point within a plane, as shown in FIG box;

[0050] 2)进行简化的circular POCS的正交投影迭代: [0050] 2) simplified orthogonal projection of a circular POCS iteration:

[0051] [0051]

Figure CN103200670AD00065

[0053] 其中 [0053] in which

Figure CN103200670AD00066

表示正交凸集投影点, It represents an orthogonal projections onto convex sets points,

Figure CN103200670AD00067

sin (φ)]Γ表示Pi到Pi+1的向量,可以通过i表示出迭代的顺序,此顺序是根据感知用户的顺序确定;当后续迭代进入死循环或者不收敛时,可改变迭代的顺序,继续迭代。 sin (φ)] Γ represents the vector Pi to Pi + 1, i can be expressed iteration order, this order is determined according to the user's perception of the sequence; when subsequent iterations not converged into an infinite loop or may change the order iteration continued iteration.

[0054] 步骤二.对步骤一中获取的Mc个迭代点,进行me步回溯检查。 [0054] Step II. Mc iteration of points acquired in a step, the step for me back check. 计算相邻迭代点之间的距离I |xm+「xml I,其中,m=Mc-l,...,Mc-mc,并与门限值λ进行比较。其中,λ的取值大小取决于感知用户对主用户测距结果的平均值,且λ相对于该平均距离是一个很小值,例如:λ与该平均距离的比值小于等于0.02,考虑到算法的运算复杂度,可将该比值进一步限定在0.005〜0.02内。 I calculate the distance between the adjacent iterates | xm + "xml I, where, m = Mc-l, ..., Mc-mc, and compared with a threshold value [lambda] where, λ depends on the value. average user perception of the ranging result of the primary user, and with respect to the average distance [lambda] is a small value, for example: λ ratio of the average distance is 0.02 or less, taking into account the computational complexity of the algorithm, the ratio can be further defined within 0.005~0.02.

[0055] 回溯检查步数为me,该me的取值为参与定位的感知用户的数目L的整数倍。 [0055] me back to check the number of steps, the value of the number of perceived me users participating in the positioning of an integer multiple of L.

[0056] 步骤三.如果步骤二中的回溯比较检查中,相邻迭代点之间的距离均小于λ且不为零,则可以判断主用户位于感知用户所形成的多边形之内,将步骤一中最后L个迭代均值,即感知用户所属各个凸集的最后一次迭代的均值 [0056] Step III. If the comparison check back in step two, the distance between adjacent iterations is less than λ and not zero point, the user can determine whether the host is located within the polygon formed by the perception of the user, a step the last iteration of the mean of L, i.e. the mean perceived last user belongs convex sets each iteration

Figure CN103200670AD00071

作为主用 As the primary

户的位置信息的定位结果;如果步骤二中的回溯比较检查中,相邻迭代点之间的距离存在部分为零或大于λ的情况,继续执行步骤四。 Positioning result of the position information of the user; if the traceback step two comparative examination, where λ is greater than zero or part of the distance between adjacent points iterations exists, proceed to Step IV.

[0057] 步骤四.以第Mc步迭代结果χΜ。 [0057] Step Four. ΧΜ iterative step results Mc. 为初始点Iv进行凸集圆域边界上正交投影迭代,迭代检查步数为Mb,得到Mb个迭代点。 Iterative orthogonal projection on a convex set as the initial point of the boundary circle domain Iv, checking the iteration step number Mb, Mb iterations to obtain points.

[0058] 设定边界正交投影迭代的检查步数为Mb,由于取消了被迭代点位置的判断,边界正交投影迭代的收敛速度是不确定的,可能会很快收敛到主用户附近,也可能陷入缓慢循环,因此Mb取较大值,使边界正交投影迭代充分。 [0058] Check the number of steps is set as the orthogonal boundary projection Mb iteration, the iteration is determined by eliminating the position of the boundary orthogonal projection iteration convergence rate is uncertain, you can quickly converge to the vicinity of the main user, can get into slow cycle, so Mb whichever is greater, so that sufficient iterations orthogonal boundary projection.

[0059] 其中,根据边界正交投影迭代的规则,对主用户进行Mb步的迭代步骤为: Iteration step [0059] wherein, according to the rules of orthogonal boundary projection iteration, the main steps for user Mb:

[0060] I)设置初始点bQ, b0 = xMc [0060] I) the initial set point bQ, b0 = xMc

[0061] 2) bh+1 = PhmodL(bh), h = 0,1, 2, 3...Mb-1 [0061] 2) bh + 1 = PhmodL (bh), 0,1, 2, 3 h = ... Mb-1

[0062]其中, [0062] wherein,

Figure CN103200670AD00072

[0063] 步骤五.对步骤四中获取的Mb个迭代点,进行回溯mb步检查。 [0063] Step V. Mb iteration of points acquired in step four, retroactive mb inspection step. 计算相邻两个迭代点间的距离值,并与门限值λ进行比较。 Calculating distance values ​​between two adjacent points iterations and compared with a threshold value λ.

[0064] 回溯检查步数为mb,该mb取值与me相同,即为L的整数倍。 [0064] The back check step number mb, mb value the same with me, L is the integer multiple.

[0065] 步骤六.如果步骤五中的回溯比较检查中,相邻迭代点间的距离值均小于入,则将步骤四中最后L个迭代均值,即感知用户所属各个凸集的最后一次迭代的均值 [0065] Step 6. If the comparison check Fifth step back, the value of the distance between adjacent points is less than the iteration, then the last L four-step iteration of the mean, i.e. each user belongs perceived last iteration Convex Sets mean

Figure CN103200670AD00073

作为主用户的位置信息的定位结果;如果步骤五中的回溯比较 As a result of the position information of the positioning of the primary user; if the comparison step 5 lookback

检查中,相邻迭代点间的距离值存在大于λ的情况,说明边界正交投影在经过了Mb步充分迭代之后,依然没有收敛到主用户位置附近,而是陷入了缓慢循环迭代。 Check the distance between the adjacent value exists iterates larger than λ, the boundary orthogonal projection described after a full iteration step Mb, still does not converge to the vicinity of the primary user's location, but slowly into a loop iteration. 此时,跳转步骤四继续执行凸集圆域边界上的正交投影迭代,其中,以前次Mb步正交投影迭代中第Mb步迭代结果V为初始点Iv变换原有的投影迭代顺序,直到相邻迭代点间的距离值均小于λ。 In this case, the step jumps four iterations continue orthogonal projection on a convex set circular region boundary, wherein the result of the previous iteration step Mb orthogonal projection views Mb iteration step V conversion original projection point as the initial iteration order Iv, until the value of the distance between adjacent points of iterations is less than λ.

[0066] 以下结合附图和具体的实例来对本发明做进一步的详细说明。 [0066] The following specific examples in conjunction with the accompanying drawings and described in detail to further the present invention.

[0067] 步骤一.以L个感知用户的坐标为圆心,利用Circular POCS对主用户进行Mc步正交投影迭代,得到Mc个迭代点xk。 [0067] Step a. In the L perceived as the center coordinates of the user, the main user using iterative orthogonal projection Mc step Circular POCS, Mc iteration points to give XK.

[0068] 设定感知用户数目为L=3,感知用户的位置坐标为[(700ml500m),(500ml000m),(1000m,1000m)]。 [0068] The setting of the number of users perceived L = 3, the user's perception of the position coordinates [(700ml500m), (500ml000m), (1000m, 1000m)]. 其中,输入白噪声作为感知用户获取的与主用户之间距离的测量值屯,方差为8m。 Wherein the measured value of the distance between the input Tun obtained as white noise perceived by the user and the primary user, the variance of 8m. 设定圆凸集投影迭代步数上线Mc=10,以坐标位置Xq= (1600m,2100m)为起始位置(如图2方块所示)进行投影迭代,得到10个迭代点xk, k=l, 2,3,...10。 Circular projection on convex sets setting the iteration number line Mc = 10, the coordinate position Xq = (1600m, 2100m) projected iteration as a starting position (shown in Figure 2 block) to give 10 iterate xk, k = l 2, 3, ... 10.

[0069] 步骤二.对步骤一中获取的10个迭代点,进行me步回溯检查,回溯检查步数为mc=2L=6。 [0069] Step two 10 of Iterative acquired in step a, step for me back check, check the number of steps back to mc = 2L = 6. 计算相邻迭代点之间的距离,并与门限值λ进行比较。 Calculating the distance between adjacent points of iterations and compared with a threshold value λ. 假设所有参与定位的感知用户所获取的与主用户之间距离的测量值的平均值为R,则检查门限值λ设定为相对R的一个较小量,这里设定λ与距离平均值的比值为λ/R=0.01。 Suppose average of the measurements of the distance between the sensing positioning of all participating users of the acquired primary user is R, the threshold value λ is set to check a relatively small amount of R is set where the average distance λ the ratio λ / R = 0.01.

[0070] II Xm+1-Xm Il ( λ,其中,m=Mc-l,..., Mc-mc [0070] II Xm + 1-Xm Il (λ, where, m = Mc-l, ..., Mc-mc

[0071] 步骤三.由于步骤二中的回溯比较检查中,相邻迭代点之间的距离存在部分为零或大于λ的情况,继续执行步骤四。 [0071] Step III. Step II since retrospective comparison check, the distance between adjacent portions iterates presence or greater than zero is λ, proceed to Step IV.

[0072] 从图2中可以看出,从初始点X。 [0072] As can be seen from Figure 2, X. from the initial point 开始经过两步Circular POCS的迭代,迭代点便停滞于三个凸集圆域的交集区域上,此时根据回溯检查判断的结果,迭代点的变化差值存在一部分为零的情况,因而继续执行步骤四,进行向凸集圆域边界上的正交投影迭代。 Iterative has elapsed, a two-step iterative point is then Circular POCS stagnation region on the intersection of three sets convex circular region, in which case the result of the determination back check, the presence of a portion of the zero point of the differential variation of iterations, thus continue step four, the orthogonal projection on a convex set of iterations round the domain boundary.

[0073] 步骤四.以第Mc步迭代结果xM。 [0073] Step IV. In the first iteration step results Mc xM. 为初始点,进行凸集圆域边界上的正交投影迭代对主用户进行Mb步正交投影迭代,得到Mb个迭代点bh,h=l, 2,3,...Mb。 The initial point, the orthogonal projection on a convex set iteration round gamut boundaries primary user Mb iteration step the orthogonal projection, to give Mb iterations points bh, h = l, 2,3, ... Mb.

[0074] 首先,设定边界正交投影迭代步数上限Mb=30,由于取消了被迭代点位置的判断,边界正交投影迭代的收敛速度是不确定的,可能会很快收敛到主用户附近,也可能陷入缓慢循环,因此Mb取较大值,使边界正交投影迭代充分;之后,根据边界正交投影迭代的规则,对主用户进行Mb步定位投影,得到10个迭代点bh,h=l, 2, 3,...30。 [0074] First, set the iteration number limit boundary orthogonal projection Mb = 30, due to the cancellation point position is determined iteration, the convergence speed of the iterative orthogonal boundary projection is uncertain, you can quickly converge to the primary user near, you can get into slow cycle, so Mb whichever is greater, so that the orthogonal projection of iterations sufficient boundary; then, according to the rules of orthogonal boundary projection iteration step Mb primary user positioning projection, to give 10 iterate BH, h = l, 2, 3, ... 30.

[0075] 步骤五.对步骤四中获取的Mb个迭代点,进行回溯mb步检查。 [0075] Step V. Mb iteration of points acquired in step four, retroactive mb inspection step. 回溯检查的步数mb=2L=6。 Checking the number of steps back mb = 2L = 6. 计算相邻两个迭代点间的距离值,并与门限值λ进行比较。 Calculating distance values ​​between two adjacent points iterations and compared with a threshold value λ.

[0076] 步骤六.在迭代了Mb步之后,回溯检查mb=6步的迭代点变化值,发现相邻迭代点间的距离值存在大于λ的情况,投影迭代陷入了缓慢循环投影迭代,因此,需要跳转到步骤四,并以首次Mb步正交投影迭代中第Mb步的迭代结果为初始点Iv再次执行凸集圆域边界上的Mb步正交投影迭代,此时变换首次的投影迭代顺序,将原投影迭代的顺序由Ρ1-Ρ2-Ρ3改为Ρ2-Ρ3-Ρ1,当再次经过Mb步迭代,回溯检查mb=6步,发现迭代点的变化值小于门限λ,说明迭代收敛主用户位置附近,因此将第二次Mb步正交投影迭代中最后L个迭代均值(如图2星号所示)确定主用户位置。 [0076] Step VI. After the iteration of step Mb, back check mb = 6 Iterative step change value, the value found from the presence of iterations between adjacent points where λ is greater than the projection into a slow loop iterations iterative projection, so requires jumps to step four, and the results of the first iterative step Mb orthogonal projection of the first iteration step the initial point Iv performed Mb Mb iteration step the orthogonal projection on a convex set gamut boundary circle again, this time the first projection transform iteration order, the original sequence is replaced by a iterative projection Ρ1-Ρ2-Ρ3 Ρ2-Ρ3-Ρ1, when a Mb again after iteration, step back check mb = 6, iterative point changes found value is less than the threshold [lambda], described iterative convergence position near the main user, so the second step the orthogonal projection Mb iteration last L iteration of the mean (asterisks in FIG. 2) determines the position of the primary user.

[0077] 图3是Hybrid POCS定位算法与BackCheck POCS定位算法的定位仿真结果比较图。 [0077] FIG. 3 is a simulation result comparing FIG positioning Hybrid POCS algorithm BackCheck POCS positioning location algorithm. 图中横坐标为仿真重复次数,纵坐标是估计位置与目标真实位置之间的差值与感知用户到主用户之间真实距离平均值的比值。 Abscissa the simulation repetitions, and the ordinate is the difference between the perception of the user's position and the target true position estimation to the ratio between the average actual distance of the primary user. 从图3中可以看出,一般情况下两种算法的定位精度比较接近,但是在某些情况下,BackCheck POCS定位算法的定位精度比较有优势。 As can be seen from Figure 3, the positioning accuracy of the two algorithms generally close to, but in some cases, the positioning accuracy of the positioning BackCheck POCS algorithm is an advantage. 这是因为,当主用户远离感知用户时,双曲线的渐近线性质容易致使双曲线的交点受到测距噪声的波动影响比较明显,因此在这种情况下,可以看出BackCheck POCS算法的定位效果要比Hybrid POCS算法优越。 This is because, when the main user away from the user perception, hyperbolic asymptotes properties easily affected by intersection of the hyperbolas causes noise fluctuations ranging obvious, so in this case, it can be seen positioned BackCheck POCS algorithm results than Hybrid POCS algorithm is superior.

[0078] 图4描述了Hybrid POCS定位算法与BackCheck POCS定位算法在不同测距误差影响下,定位精度的比较。 [0078] FIG 4 depicts a comparison positioned Hybrid POCS algorithm BackCheck POCS algorithm positioned at different distance errors on the positioning accuracy. 从图4中可以看出,BackCheck POCS定位算法比Hybrid POCS定位算法具有一定的优势,这主要是由于Hybrid POCS算法中双曲线投影定位对于主用户在感知用户多边形之外的情况下收敛点受噪声波动较大,因此随着测距误差的增大。 As can be seen from Figure 4, BackCheck POCS localization algorithm has certain advantages over Hybrid POCS positioning algorithm, which is mainly due to Hybrid POCS algorithm hyperbolic positioning projection for the primary user perception of the user other than the case of the convergence point of the polygon by noise fluctuations, so as the ranging error is increased.

[0079] 尽管已经示出和描述了本发明的实施例,对于本领域的普通技术人员而言,可以理解在不脱离本发明的原理和精神的情况下可以对这些实施例进行多种变化、修改、替换和变型,本发明的范围由所附权利要求及其等同限定。 [0079] While there has been illustrated and described embodiments of the present invention, those of ordinary skill in the art, to be understood that various changes may be made to these embodiments without departing from the principles and spirit of the present invention, modifications, substitutions and modifications, the scope of the invention being indicated by the appended claims and their equivalents.

Claims (5)

1.一种认知无线电网络中对主用户定位的方法,其特征在于,该方法包括以下步骤: 步骤一、以L个感知用户的坐标为圆心,利用凸集投影算法对主用户进行Mc步正交投影迭代,得到Mc个迭代点χ k,其中k = 1,2,3,...,Mc ; 步骤二.对步骤一中获取的Mc个迭代点,进行me步回溯比较检查,计算出相邻迭代点之间的距离I |xm+「xml I,其中,m=Mc-l,...,Mc-mc,并与门限值λ进行比较;步骤三.如果步骤二中的回溯比较检查中,相邻迭代点之间的距离均小于λ且不为零,则将步骤一中最后L个迭代均值作为主用户位置信息的定位结果;如果步骤二中的回溯比较检查中,相邻迭代点之间的距离存在部分为零或大于λ的情况,继续执行步骤四;步骤四.以Mc步迭代结果χΜ。为初始点Iv进行凸集圆域边界上正交投影迭代,迭代检查步数为Mb,得至IJ Mb个迭代点bh,其中h=l, 2,3,..., CLAIMS 1. A method for cognitive radio networks primary user location, characterized in that the method comprises the following steps: a step to the user's perception of L as the center coordinates, using the POCS algorithm steps primary user Mc iterative orthogonal projection, to give Mc iteration points χ k, where k = 1,2,3, ..., Mc; Mc iteration step two pairs of points obtained in one step, for comparison check me back step, calculation. the distance I between the adjacent iterates | xm + "xml I, where, m = Mc-l, ..., Mc-mc, and compared with a threshold value [lambda]; if back in step three step two. comparison inspection, the distance between adjacent points are less than λ iterative and non-zero, then a step in the iteration of the mean as the last L positioning results primary user location information; if back comparison inspection in step two, the phase there is a distance between neighboring partial iterates zero or greater than λ, and proceed to step four;. step four sets for convex gamut boundary circle orthogonal projection iteration, the iteration check point Iv initial iteration results in Mc χΜ step number Mb, to obtain IJ Mb iterations points bh, where h = l, 2,3, ..., Mb ; 步骤五.对步骤四中获取的Mb个迭代点,进行mb步回溯比较检查,计算相邻两个迭代点间的距离值I |bn+1-bn| I,其中,n=Mb-l,...,Mb-mb,并与门限值λ进行比较; 步骤六.如果步骤五中的回溯比较检查中,相邻迭代点间的距离值均小于λ,则将步骤四中最后L个迭代均值作为主用户位置信息的定位结果;如果步骤五中的回溯比较检查中,相邻迭代点间的距离值存在大于λ的情况,跳转到步骤四以Mb步迭代结果b-为初始点Iv并变换投影迭代顺序,直到相邻迭代点间的距离值均小于λ。 . Mb; Mb iteration step five pairs of points obtained in step four, for comparison check mb step back, I calculated distance value between adjacent two iterates | bn + 1-bn | I, where, n = Mb- l, ..., Mb-mb, and a threshold value [lambda]; step six check if the fifth step back in comparison, the value of the distance between adjacent points of iterations is less than [lambda], then the last four steps. L iterations the mean position location as the primary user of the result information; check if the comparison step back Fifth, the present value of the distance between adjacent points is larger than λ iteration, the four jumps to step iteration in Mb is the result in step b- and converting the initial point Iv projected iteration order, until the value of the distance between adjacent points of iterations is less than λ.
2.如权利要求1所述的认知无线电网络中对主用户定位的方法,其特征在于,所述步骤一包括: 1.1)初始化步骤:设置初始点Χο,其中X。 2. The method of claim 1 cognitive radio network user location in the main claim, wherein said step a comprises: 1.1) initialization step: the initial set point Χο, wherein X. 为任意位置上的一点; 1.2)利用以下公式进行投影迭代: To a point on an arbitrary position; 1.2) using the following equation projection iteration:
Figure CN103200670AC00021
其中,PjDi(Xfc)表示正交凸集投影点,dJcosOjo),sin (φ)]τ表示Pi到Pi+1的向量! Wherein, PjDi (Xfc) represents an orthogonal projection onto convex sets points, dJcosOjo), sin (φ)] τ represents the vector of Pi to Pi + 1! Pi为第i个感知用户的位置坐标,ie [i,l]。 Pi is the i-th user's perceived position coordinate, ie [i, l]. Di为以第i个感知用户为圆心,以第i个感知用户测得的与主用户之间的距离测量值为半径的凸集圆域。 Di is the i-th to user perceived as the center, the distance between the measured value to the radius of an i-th user perceived measured round convex set primary user domain.
3.如权利要求1所述的认知无线电网络中对主用户定位的方法,其特征在于,λ的取值大小取决于感知用户对主用户测距结果的平均值,且λ相对于该平均距离是一个很小值。 3. The method of claim 1 cognitive radio network user location in the main claim, characterized in that, the value of [lambda] depends on the user perception of the average distance to the primary user of the result, and [lambda] with respect to the mean distance is a small value.
4.如权利要求1所述的认知无线电网络中对主用户定位的方法,其特征在于,mb与me取值相同,为L的整数倍。 4. The method of claim 1 cognitive radio network user location in the main claim, characterized in that, with the same mb me value, an integer multiple of L.
5.如权利要求1所述的认知无线电网络中对主用户定位的方法,其特征在于,所述步骤四包括: .4.1)初始化步骤:1)设置初始点bQ, bQ = xMc .4.2)利用以下公式进行投影迭代: bh+1 = PhmodL (bh),h = 0,1,2,3...Mb-1 其中, 5. The method of claim 1 cognitive radio network user location in the main claim, wherein said four steps comprises: .4.1) Initialization steps: 1) Set the initial point bQ, bQ = xMc .4.2) using the following iterative formula projection: bh + 1 = PhmodL (bh), h = 0,1,2,3 ... Mb-1 wherein,
Figure CN103200670AC00022
其中,Pi为第i个感知用户的位置坐标,i ∈ [1,L]。 Wherein, Pi is the i-th user's perceived position coordinates, i ∈ [1, L]. di是第i个感知用户测得的与主用户之间的距离测量值; di is the distance measurements between the i th user perception of the measured primary user;
Figure CN103200670AC00031
是第i个感知用户所确定的半径为Cli的圆边界。 Sensing the i-th user is determined by the radius of the boundary circle Cli.
CN 201310057842 2013-02-25 2013-02-25 Check back POCS based cognitive radio primary user positioning method CN103200670B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN 201310057842 CN103200670B (en) 2013-02-25 2013-02-25 Check back POCS based cognitive radio primary user positioning method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN 201310057842 CN103200670B (en) 2013-02-25 2013-02-25 Check back POCS based cognitive radio primary user positioning method

Publications (2)

Publication Number Publication Date
CN103200670A true true CN103200670A (en) 2013-07-10
CN103200670B CN103200670B (en) 2015-08-05

Family

ID=48722986

Family Applications (1)

Application Number Title Priority Date Filing Date
CN 201310057842 CN103200670B (en) 2013-02-25 2013-02-25 Check back POCS based cognitive radio primary user positioning method

Country Status (1)

Country Link
CN (1) CN103200670B (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104320845A (en) * 2014-07-04 2015-01-28 南京邮电大学 A main user positioning method based on sensor and quantum intelligent computing

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US8155673B2 (en) * 2006-05-08 2012-04-10 Skyhook Wireless, Inc. Estimation of position using WLAN access point radio propagation characteristics in a WLAN positioning system
CN102752849A (en) * 2012-02-29 2012-10-24 中国人民解放军理工大学 Single receiving machine location method based on signal detection probability and wave angle estimation

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US8155673B2 (en) * 2006-05-08 2012-04-10 Skyhook Wireless, Inc. Estimation of position using WLAN access point radio propagation characteristics in a WLAN positioning system
CN102752849A (en) * 2012-02-29 2012-10-24 中国人民解放军理工大学 Single receiving machine location method based on signal detection probability and wave angle estimation

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
于浩等: "加权凸集投影法在分布式目标定位中的应用", <<计算机工程>>, 30 June 2011 (2011-06-30) *

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104320845A (en) * 2014-07-04 2015-01-28 南京邮电大学 A main user positioning method based on sensor and quantum intelligent computing

Also Published As

Publication number Publication date Type
CN103200670B (en) 2015-08-05 grant

Similar Documents

Publication Publication Date Title
Blatt et al. Energy-based sensor network source localization via projection onto convex sets
CN101778399A (en) Method for optimizing WLAN (Wireless Local Area Network) indoor ANN (Artificial Neural Network) positioning based on FCM (fuzzy C-mean) and least-squares curve surface fitting methods
Sanghavi et al. Message passing for maximum weight independent set
CN101751695A (en) Estimating method of main curvature and main direction of point cloud data
Zhou et al. A context sensitive offloading scheme for mobile cloud computing service
Bansal et al. A polylogarithmic-competitive algorithm for the k-server problem
Jia et al. A set-theoretic approach to collaborative position location for wireless networks
Son et al. The value of clustering in distributed estimation for sensor networks
US20060165052A1 (en) Approach for determining the real time availability of a group of network elements
Demaine et al. Minimizing the diameter of a network using shortcut edges
CN101620270A (en) Wireless location method based on cluster-fusion
Zhang et al. To offload or not to offload: an efficient code partition algorithm for mobile cloud computing
US20140143407A1 (en) Multi-objective server placement determination
CN102364899A (en) Particle-swam-optimization-based virtual network mapping method and system
Goussevskaia et al. Algorithms for wireless capacity
CN101754226A (en) Reconstruction method of terminal under environment of cognitive radio network
Wang et al. Device-free localization with multidimensional wireless link information
CN101251592A (en) Method for locating node of wireless sensor network
CN102186176A (en) Cognitive radio spectrum sharing method based on supply-demand balance
Benouaret et al. Selecting skyline web services from uncertain qos
Kalathil et al. Spectrum sharing through contracts for cognitive radios
CN101848012A (en) Perception method of cooperative spectrum
Razaviyayn Successive convex approximation: Analysis and applications
CN101778005A (en) Method and system for configuring complex network
Luke Local linear convergence of approximate projections onto regularized sets

Legal Events

Date Code Title Description
C06 Publication
C10 Entry into substantive examination
C14 Grant of patent or utility model