802.11ac receiver minimum input sensitivity test -凯发k8网页登录

this example shows how to measure the receiver minimum input sensitivity as specified in section 21.3.18.1 of the ieee std 802.11™-2020 standard [ 1 ].

introduction

the receiver minimum sensitivity test ensures a device under test (dut) receives data with a defined maximum packet error rate (per) of 10% at a defined minimum signal power. the minimum signal power depends on the channel bandwidth and modulation and coding scheme (mcs) as specified in table 21-25 of ieee std 802.11™-2020 [ 1 ]:

when the test is performed with hardware, each input antenna port on the dut is connected through a cable to a single output antenna port of a transmitter. to perform the test, specify these parameters for the test waveform:

number of spatial streams - equal to the number of transmit antennas
psdu length, in bytes - 4096
space-time block coding (stbc) - disabled
guard interval, in nanoseconds - 800
channel coding - binary convolutional coding (bcc)

this example shows how to simulate the test by using wlan toolbox™. vht packets stimulate a receiver at a range of input levels below the minimum sensitivity level. the example then measures the packet error rate for each sensitivity level.

the example simulates the test by performing these steps over a range of sensitivity levels:

generate and scale packets to the desired signal level
add white gaussian noise is to create a noise floor at the receiver
demodulate the noisy packets to recover psdus.
compare recovered psdus to those transmitted to determine the number of packet errors and hence the packet error rate.

automatic gain control (agc), packet detection, timing synchronization, carrier frequency offset correction, noise estimation and phase tracking are performed by the example receiver. this diagram demonstrates processing for each packet:

test parameters

configure a transmission for the test by using a vht configuration object. this example measures the minimum sensitivity for a 160 mhz transmission with 64-qam rate 5/6 modulation and coding. the simulated dut has 2 receive antennas. test different configurations by changing these parameters.

cfgvht = wlanvhtconfig;             % create vht transmission configuration
cfgvht.channelbandwidth = 'cbw160'; % bandwidth
cfgvht.mcs = 7;                     % 64-qam, rate 5/6
numreceiveantennas = 2;             % number of receive antennas

the test requires these fixed transmission parameters.

cfgvht.apeplength = 4096; % bytes
cfgvht.stbc = false;
cfgvht.numtransmitantennas = numreceiveantennas;
cfgvht.numspacetimestreams = numreceiveantennas;
cfgvht.spatialmapping = 'direct';
cfgvht.guardinterval = 'long';

simulation parameters

a receiver processes vht packets at a range of input levels below the minimum input sensitivity level. specify the range of offsets to test in the vector testinputleveloffsets.

testinputleveloffsets = [-10 -9 -8 -7]; % db

control the number of packets tested at each sensitivity by specifying these parameters:

maxnumerrors is the maximum number of packet errors simulated at each input level. when the number of packet errors reaches this limit, the simulation at this sensitivity level is complete.
maxnumpackets is the maximum number of packets simulated at each input level and limits the length of the simulation if the packet error limit is not reached.

the numbers chosen in this example lead to a very short simulation. increase maxnumerrors and maxnumpackets for meaningful results.

maxnumerrors = 20;
maxnumpackets = 200;

signal power setup

the minimum sensitivity test specifies a maximum per for a measured input level per receive antenna. in this simulation the receiver processes a test signal with a specified input level in dbm. generate the test signal using the function. the function normalizes the waveform such that the power for all antennas sums to 0 dbm. therefore, scale the output of the waveform generator to create the desired input level.

% receiver minimum input level sensitivity for 20 mhz, table 21-25. the
% sensitivity increases by 3db for double the bandwidth.
rxminsensitivitytable = [-82 -79 -77 -74 -70 -66 -65 -64 -59 -57]; % dbm
% get minimum input sensitivity given mcs and bandwidth
fs = wlansamplerate(cfgvht);  % baseband sampling rate (hz)
b = floor(10*log10((fs/20e6))); % scalar for bandwidth
rxminsensitivity = rxminsensitivitytable(cfgvht.mcs 1) b; % dbm
disp(['minimum sensitivity for mcs' num2str(cfgvht.mcs) ', ' ...
    num2str(fs/1e6) ' mhz: ' num2str(rxminsensitivity,'%2.1f') ' dbm'])

minimum sensitivity for mcs7, 160 mhz: -55.0 dbm

define the range of input levels below the minimum level to test using testinputlevels.

testinputlevels = rxminsensitivity testinputleveloffsets; % dbm

calculate a voltage scalar, a, to scale the generated waveform for each test level. the power per receive antenna port is measured during the simulation to confirm the input signal level is correct.

a = 10.^((testinputlevels-30)/20);      % voltage gain (attenuation)
a = a*sqrt(cfgvht.numtransmitantennas); % account for generator scaling

noise configuration

add thermal noise at the receiver. the height of the noise floor determines the snr at the receiver, as the input signal level is fixed for this test. the noise figure of the receiver determines the level of noise floor.

nf = 6;         % noise figure (db)
t = 290;        % ambient temperature (k)
bw = fs;        % bandwidth (hz)
k = 1.3806e-23; % boltzmann constant (j/k)
noisefloor = 10*log10(k*t*bw) nf; % db
disp(['receiver noise floor: ' num2str(noisefloor 30,'%2.1f') ' dbm'])

receiver noise floor: -85.9 dbm

add noise to the waveform using an awgn channel, .

awgnchannel = comm.awgnchannel('noisemethod','variance', ...
    'variance',10^(noisefloor/10));

input level sensitivity simulation

calculate the packet error rate for each input level by simulating multiple packets.

for each packet perform the following processing steps:

create and encode a psdu to create a single packet waveform.
create the desired input level in dbm by scaling the waveform.
measure the power of the received waveform.
add awgn to the received waveform.
boost the signal prior to processing by passing through an automatic gain control.
detect the packet.
estimate and correct coarse carrier frequency offset.
establish fine timing synchronization.
estimate and correct fine carrier frequency offset.
extract and ofdm demodulate the vht-ltf and perform channel estimation.
extract the vht data field and recover the psdu.

ind = wlanfieldindices(cfgvht); % for accessing fields within the packet
chanbw = cfgvht.channelbandwidth;
rng(0);  % set random state for repeatability
agc = comm.agc; % automatic gain control
s = numel(testinputlevels);
packeterrorrate = zeros(s,1);
rxantennapower = zeros(s,1);
for i=1:s
    disp(['simulating ' num2str(testinputlevels(i),'%2.1f') ...
        ' dbm input level...']);
    % loop to simulate multiple packets
    numpacketerrors = 0;
    measuredpower = zeros(maxnumpackets,1); % average power per antenna
    numpkt = 1; % index of packet transmitted
    while numpacketerrors<=maxnumerrors && numpkt<=maxnumpackets
        % generate a packet waveform
        txpsdu = randi([0 1],cfgvht.psdulength*8,1); % psdulength in bytes
        tx = wlanwaveformgenerator(txpsdu,cfgvht);
        % scale input signal to desired level
        rx = tx.*a(i);
        % measure the average power at the antenna connector in watts
        measuredpower(numpkt) = mean(mean(rx.*conj(rx)));
        % add noise floor at receiver
        rx = awgnchannel(rx);
        % pass each channel through agc
        for ic = 1:size(rx,2)
            rx(:,ic) = agc(rx(:,ic));
            reset(agc);
        end
        % packet detect and determine coarse packet offset
        coarsepktoffset = wlanpacketdetect(rx,chanbw);
        if isempty(coarsepktoffset) % if empty no l-stf detected; packet error
            numpacketerrors = numpacketerrors 1;
            numpkt = numpkt 1;
            continue; % go to next loop iteration
        end
        % extract l-stf and perform coarse frequency offset correction
        lstf = rx(coarsepktoffset (ind.lstf(1):ind.lstf(2)),:);
        coarsefreqoff = wlancoarsecfoestimate(lstf,chanbw);
        rx = frequencyoffset(rx,fs,-coarsefreqoff);
        % extract the non-ht fields and determine fine packet offset
        nonhtfields = rx(coarsepktoffset (ind.lstf(1):ind.lsig(2)),:);
        finepktoffset = wlansymboltimingestimate(nonhtfields,chanbw);
        % determine final packet offset
        pktoffset = coarsepktoffset finepktoffset;
        % if packet detected out of a reasonable range (>50 samples);
        % packet error
        if pktoffset>50
            numpacketerrors = numpacketerrors 1;
            numpkt = numpkt 1;
            continue; % go to next loop iteration
        end
        % extract l-ltf and perform fine frequency offset correction
        lltf = rx(pktoffset (ind.lltf(1):ind.lltf(2)),:);
        finefreqoff = wlanfinecfoestimate(lltf,chanbw);
        rx = frequencyoffset(rx,fs,-finefreqoff);
        % extract vht-ltf samples from the waveform, demodulate and perform
        % channel estimation
        vhtltf = rx(pktoffset (ind.vhtltf(1):ind.vhtltf(2)),:);
        vhtltfdemod = wlanvhtltfdemodulate(vhtltf,cfgvht);
        [chanest,chanestsspilots] = wlanvhtltfchannelestimate(vhtltfdemod,cfgvht);
        % extract vht data samples from the waveform
        vhtdata = rx(pktoffset (ind.vhtdata(1):ind.vhtdata(2)),:);
        % estimate the noise power in vht data field
        nestvht = vhtnoiseestimate(vhtdata,chanestsspilots,cfgvht);
        % recover the transmitted psdu in vht data
        rxpsdu = wlanvhtdatarecover(vhtdata,chanest,nestvht,cfgvht, ...
            'ldpcdecodingmethod','norm-min-sum');
        % determine if any bits are in error, i.e. a packet error
        packeterror = any(biterr(txpsdu,rxpsdu));
        numpacketerrors = numpacketerrors packeterror;
        numpkt = numpkt 1;
    end
    % calculate packet error rate (per) at input level point
    packeterrorrate(i) = numpacketerrors/(numpkt-1);
    disp(['  completed after ' ...
        num2str(numpkt-1) ' packets, per: ' ...
        num2str(packeterrorrate(i))]);
    % calculate average input power per antenna
    rxantennapower(i) = 10*log10(mean(measuredpower(1:(numpkt-1)))) 30;
    disp(['  measured antenna connector power: ' ...
        num2str(rxantennapower(i),'%2.1f') ' dbm']);
end

simulating -65.0 dbm input level...
  completed after 21 packets, per: 1
  measured antenna connector power: -65.0 dbm
simulating -64.0 dbm input level...
  completed after 26 packets, per: 0.80769
  measured antenna connector power: -64.0 dbm
simulating -63.0 dbm input level...
  completed after 130 packets, per: 0.16154
  measured antenna connector power: -63.0 dbm
simulating -62.0 dbm input level...
  completed after 200 packets, per: 0.02
  measured antenna connector power: -62.0 dbm

analysis and further exploration

plot the per for tested input signal levels with the maximum per at minimum sensitivity.

figure
semilogy(rxantennapower,packeterrorrate,'o-')
hold on
semilogy(rxminsensitivity,0.1,'rx')
currentxlim = xlim(gca);
xlim([currentxlim(1) currentxlim(2) 1])
grid on
xlabel('measured power per antenna connector (dbm)');
ylabel('per');
legend('simulated per performance','maximum per at minimum sensitivity');
title(sprintf(['minimum input sensitivity test: mcs%d, %d mhz, ' ...
    '%d antennas'],cfgvht.mcs,fs/1e6,cfgvht.numtransmitantennas))

the plot reveals the simulated 10% per is just under 8 db lower than the minimum sensitivity specified by the test. this difference is due to the implementation margin allowed by the test. the implementation margin allows for algorithmic degradations due to impairments and the receiver noise figure when compared to ideal awgn performance [ 2 ]. in this example only awgn is added as an impairment. therefore, only the algorithmic performance of front-end synchronization, channel estimation and phase tracking in the presence of awgn use the implementation margin. if more impairments are included in the simulation the per waterfall in the plot will move right towards the minimum sensitivity and the margin will decrease.

the number of packets tested at each snr point is controlled by two parameters; maxnumerrors and maxnumpackets. for meaningful results, you should use larger numbers than those used in this example.

selected bibliography

ieee std 802.11™-2020. ieee standard for information technology - telecommunications and information exchange between systems - local and metropolitan area networks - specific requirements - part 11: wireless lan medium access control (mac) and physical layer (phy) specifications.
perahia, eldad, and robert stacey. next generation wireless lans: 802.11n and 802.11ac. cambridge university press, 2013.