无线网络定位算法的程序。TOARSSI什么算法的都可以，想找个程序作个参考。

我有matlab的。

太多了，我的QQ：39400877

%|

%| SCRIPT: simMLE

%|

%| PURPOSE: Simulate a relative location system by generating

%|random measurements and maximizing the likelihood fcn.

%|After many trials, show the results vs. the Cramer-Rao Bound.

%|

%| AUTHOR: Neal Patwari

%| http://www.engin.umich.edu/~npatwari/

%|

%| REFERENCE: Relative Location Estimation in Wireless Sensor Networks

%| (N. Patwari, A. O. Hero, M. Perkins, N. S. Correal, R. J. O'Dea),

%| IEEE Trans. Signal Processing, vol. 51, no. 8, Aug. 2003, pp. 2137-2148.

%|

tic

% Use globals to allow minimization functions access to network info,

% debugging info.

global refDevices blindDevices totalDevices linearRefLocs dhat funcEvals dfuncEvals

% Basic simulation parameters

roomSize= [1,1] % Room size, meters

gridSize= 5 % How many sensors per side

refDevices = 4 % How many references (must be same length as actualRefLocs)

trials = 20 % How many indep trials to run

measMethod = 'R'% Use 'R' for RSS, 'T' for TOA

totalDevices= gridSize^2

blindDevices= totalDevices - refDevices

blindCoords = 2*blindDevices

actualRefLocs = [0,00,11,11,0]

linearRefLocs = [actualRefLocs(:,1)', actualRefLocs(:,2)']

% Optimization parameters

ftol = 0.00001

if measMethod == 'R',

func = 'calcError' % Use for RSS

dfunc = 'calcDError' % Use for RSS

else

func = 'calcErrorTOA' % Use for TOA

dfunc = 'calcDErrorTOA' % Use for TOA

end

%| 1. Set up the blindfolded device locations

delta= 1/(gridSize-1)

coords = 0:delta:1

xMatrix = ones(gridSize,1)*coords

yMatrix = xMatrix'

xBlind = [xMatrix(2:gridSize-1), ...

xMatrix(gridSize+1:totalDevices-gridSize), ...

xMatrix(totalDevices-gridSize+2:totalDevices-1)]

yBlind = [yMatrix(2:gridSize-1), ...

yMatrix(gridSize+1:totalDevices-gridSize), ...

yMatrix(totalDevices-gridSize+2:totalDevices-1)]

actualBlindLocs = [xBlind', yBlind']

actualAllLocs = [actualRefLocsactualBlindLocs]

xActual = actualAllLocs(:,1)'

yActual = actualAllLocs(:,2)'

actualDist = L2_distance(actualAllLocs', actualAllLocs',0)

%| 2. Define the channel model

if measMethod == 'R'

sigmaOverN = 1.7

% If C==1, then this simulation runs the _true_ MLE.

% If C==exp( 0.5* (log(10)/10 *sigmaOverN)^2), then this runs a

% bias-corrected (pseudo-) MLE.

% C = exp( 0.5* (log(10)/10 *sigmaOverN)^2)

C = 1

else

sigma_d = 0.2 % Use for TOA

end

for trial = 1:trials,

if measMethod == 'R'

%| 3.0 Generate a random set of RSS-based distance measurements. When RSS

%| is expressed in dB, errors are Gaussian. Here, dhat is an interim

%| variable which has units of distance, and represents an estimate for

%| the range. It is correctly randomly generated as follows:

dhat = actualDist.*10.^(sigmaOverN/10 .*symrandn(totalDevices))./C

else

%| 3.1 Generate a set of TOA measurements, which are Gaussian around the

%| true value with variance sigma_d.

dhat = actualDist + sigma_d .* symrandn(totalDevices)

end

%| 4. Make an initial guess of the coordinates.

blindLocs0 = [xBlind, yBlind]% Use the true coordinates (unrealistic but best case)

%| 5. Find optimum locations of neurfons (fixed and relative)

funcEvals = 0 dfuncEvals = 0

[coordsMLE, iter, errorMin] = frprmn(blindLocs0, ftol, func, dfunc, 0)

disp(sprintf('%d: Function / Deriv. evals: %d / %d.', trial, funcEvals, dfuncEvals))

%| 6. Save the resulting estimated coords

coordEsts(trial, 1:blindCoords) = coordsMLE

end % for trial

estMean = mean(coordEsts)

estCov = cov(coordEsts)

estVars = diag(estCov)

estStds = sqrt(estVars)

locVars = estVars(1:blindDevices) + estVars((blindDevices+1):(2*blindDevices))

locStd = sqrt(locVars)

toc % show time of execution

% Plot the location estimates for sensors, one at a time.

if 0,

figure

for i=1:blindDevices,

clf

plot(coordEsts(:,i), coordEsts(:,blindDevices+i),'.', ...

estMean(i), estMean(blindDevices+i), 'ro')

hold on

set(gca,'xlim',[-0.2 1.2])

set(gca,'ylim',[-0.2 1.2])

set(gca,'FontSize',20)

set(gca,'DataAspectRatio',[1 1 1])

xlabel('X Position (m)')

ylabel('Y Position (m)')

set(gca,'xTick',0:0.25:1)

set(gca,'yTick',0:0.25:1)

grid

pause

end

end

% Calculate and plot CRB vs. estimator performance.

figureclf

if measMethod == 'R'

[locstdCRB, coordCRB] = calcLocalizationCRB('R', [xBlind, actualRefLocs(:,1)'], ...

[yBlind, actualRefLocs(:,2)'], blindDevices, totalDevices, sigmaOverN)

else

[locstdCRB, coordCRB] = calcLocalizationCRB('T', [xBlind, actualRefLocs(:,1)'], ...

[yBlind, actualRefLocs(:,2)'], blindDevices, totalDevices, sigma_d)

end

for i=1:blindDevices,

hold on

R = cov(coordEsts(:,i), coordEsts(:,blindDevices+i))

drawOval(estMean(i), estMean(blindDevices+i), R, 'k-','v', 8, 0, 1)

R_CRB = coordCRB([i, i+blindDevices],[i, i+blindDevices])

drawOval(xBlind(i), yBlind(i), R_CRB, 'r--','.',20, 0, 1)

end

set(gca,'xlim',[-0.2 1.2])

set(gca,'ylim',[-0.2 1.2])

set(gca,'FontSize',18)

set(gca,'DataAspectRatio',[1 1 1])

xlabel('X Position (m)')

ylabel('Y Position (m)')

set(gca,'xTick',0:0.25:1)

set(gca,'yTick',0:0.25:1)

grid

% Use for comparison

RMS_est_Std = sqrt(mean(locStd.^2))

RMS_crb_Std = sqrt(mean(locstdCRB.^2))

WIFI定位有很多种方法，比如测信号强度(RSS)，信号角度(AOA)，相位(POA)，时间(TOA)和时间差(DTOA)。

你说的通过测距离定位的方法一般是除了AOA以外的四种。

现在常用的是信号强度(RSS)检测，因为适用范围广，检测你所接受到的许多WIFI接入点(AP)的信号强度，根据信号强度确定位置。这种方法精度很低，而且应用时会有严重的逻辑错误。现在有通过事先测好的信号强度地图(radio map)来辅助定位，能够提高精度。另外也可以由其他信号比如GSM，GPRS，3G，蓝牙，红外，ZigBee等等很多种信号来辅助定位。

具体有什么算法我不细说了，因为影响定位的变量太多，算法也很多。

测时间和时间差的算法，基本原理如一楼所说，目前看来精度是最低的。。因为wifi定位肯定是用于室内的（在室外GPS的定位精度非常高，可以满足测绘需求，当然手机之类的GPS模块就不说了）在室内多路径效应严重，环境场经常变化，wifi接收机和发射机时间不容易同步或者同步精度不高（影响时间差精度），所以对定位精度的影响也很大很难建模。

有什么问题可以和我联系

你是用超宽带基站作人员定位使用么？UWB基站间的部署距离受基站的覆盖范围影响、部署环境影响。

具体数值：在部署环境比较简单，没有遮挡的情况下，做人员定位，微能信息科技建议超宽带基站间的部署距离可以调到50米的范围。

---