 Good afternoon everyone. Thanks first of all for giving me this opportunity and during this talk I will illustrate how in the polar bear experiment we measure B-mods. As you can see you will find many things in common with the of the lesson of yesterday by Bruce Partridge and I will highlight how in polar bear we deal with all the problems that were highlight and yesterday. So this is I'm Davide Polletti working at APC as a PhD student and this talk is given on the behalf of the polar bear collaboration which is Worldwide collaboration involving many institutional from all around the world mainly from the US Japan and from a con with a contribution also from Europe mainly in data analysis from APC and CISA here and also in Pirah College So this is the outline of the talk. First of all I will go through the basics of the B-mod science and measurements then I will focus on the polar bear experiment and show you the results of the first season and then conclude with the perspectives of the experiment. As we heard many times during the school the CNB is a primary source of information for cosmology however the scientific potential of the temperature and isotropy has been already fully nearly fully exploited by several generations of experiments. So we are now moving towards polarization both for improving those constraints but also for getting complementary constraints that we cannot get at all from from temperature. So what we actually measure from the from the CNB polarization are the Q and U stocks parameter that are then decomposing to a gradient like and curled like pattern the E-mod and B-mod and here on the right you see the power spectrum of these three components temperature and the two polarization and you can see how the temperature is way higher than the polarization signal and the E-mods in turn are higher than the B-mods signal and therefore measuring the B-mods is not just a matter of achieving this sensitivity but also preventing the temperature and E-mod power from leaking into your B-mod signal and this plot also show the distinction between the primordial B-mods which dominate at large scales and lens in B-mod dominating at small scales. The primordial B-mods are related to the gravitational waves at the last scattering surface because the scalar perturbations produce basically only E-type of polarization and therefore B-mods are a unique channel to probe those perturbations and being able to measure the amplitude of those perturbations the last scattering surface will then able to set an energy scale of the inflation and while measuring the tilt of the spectrum of the tensor perturbation would enable to probe the consistency relation that many inflationary models predict and So what about the lens in B-mods? During their journey towards us the CMB photons gets deflected by the gravitational lensing exerted by intervening structure here and the net effect is a small displacement that you can see there between the CMB picture that we get and what the actual last scattering surface looked like and on the harmonic domain the effect is a mixing a reshuffling of power along angular scales between the measure the E-mod Between the measure the power and what was the original power we have an L prime multiple contributing to an n-multiple, but also we have the E E-type of polarization Contributing to the B-mod polarization the actual B-mod polarization measured and those are the lensing B-mods and We can exploit this correlation in the actual CMB in the CMB picture to infer this displacement field and I Alight that this displacement field is related to this gravitational potential integrated over the line of sight and therefore it is sensitive to Every parameter that can affect the structure formations most notably the total mass of the neutrinos or the dark energy properties here I show the B-mod level lensing B-mod level for different dark energy scenarios and So what's the status of the measurement? This is the status as of March 2014. We had only we had only upper limits and The actual constraint on our didn't came from polarization measurement, but from temperature measurement However, we had major improvement during this year the 10 March the polar back collaboration published our results at small scales Bicep published the result at a large scale Act pole at small scales again The joint collaboration of bicep 2 Kekare and Planck clean the bicep point from dust and We got a couple of months ago the SPT pole papers so you see that the field is very active and These are the Direct measurement of B-mod power in the maps, but actually there are much more measurements Through cross-correlation by several experiment So the field is very active and all these experiments and others are upgrading their their instruments, so Many improvement there will there will be many improvement in the future Let's now focus on the polar bear experiment polar bear is CMB B-mod experiment Located in the Atacama Desert in Chile the peculiarity of the location is a very big sky Unaccess to a very big fraction of sky and Also the dryness of the atmosphere that makes makes it very clean in the microwaves and the target of the experiment are both the Lensing and the primordial part of the b-mod spectrum However, for the first season we focused only on very small patch deeply integrated to get a detection This is the design of the experiment as you can see the light comes Here on the primary mirror gets reflected on the second secondary enters the cryostat Here it gets modulated by a rotating helpful plate and Is focused on the focal plane the focal plane is made of more than 12 hundred volumetres Here you can see a picture of a pair of volumetres You can see here the volumetres which is sensitive to all the power Collected by these are directional antennas in front of it and since those two volumetres are sensitive are coupled coupled to those orthogonal antennas they are sensitive to different different linear combination of the stock parameters and Most notably if you take the sum and the difference of the two Measurements you can isolate the temperature part and the polarization part of the measurement and I emphasize here that depends on temperature because as the telescope scans the sky those volumetres are read out 100s of time per second making the so-called time streams of data So then we characterize our instrument using both ground-based and astrophysical Calibrators we use Jupiter for a characterizer characterizing our beam properties and that way in the crab nebula for Assessing the orientation of our volumetres once projected on the sky and Here you can see in this plot of the power spectrum of a time stream how By getting the difference of the two volumetres we can get rid of this huge contribution in noise due to the atmosphere fluctuations so once we have those calibrated time stream we are now ready to build maps and extract power spectrum from from those maps in polar better We have two pipelines for doing this we have a first pipeline that which doesn't correct in the map domain for Time domain filtering But do it directly in the power spectrum domain and the other one which corrects already for time domain filtering at map level and Having these two pipeline provides a way of cross-checking and validating the results delivered by our tools and So now we have those maps and we are ready to Probe in some science with them I Will show you now the results of the first season as I mentioned earlier the From the polarization maps we can extract an estimation of the displacement field From which we can compute the power spectrum and this is what was done in this paper in which we found a 4.2 sigma evidence of b-mods and I emphasize that this was the first measurement of of Lensing b-mods using CMB polarization data alone and Of course, we can build also this CAPA estimator and it traces the the gravitational potential along the line of sight and Of course that we can find also other tracers of this gravitational potential like the cosmic infrared background and Therefore we expect to have some correlation between this CAPA field and these other tracers That's the work done in these other paper in which we correlated the CAPA estimator of polar bear with the CAB map of from urshell and found a phone point four point two sigma evidence of polar polarisation Lensing and 2.5 sigma using estimator of CAPA field related to b-mods Of course, finally one can compute the total power b-mod power in our maps This is what's done in these other paper and those are the points measure measured from our maps this led to a 97 percent confidence level evidence of b-mods and I emphasize that as I mentioned before that the big deal is not only achieving the the sensitivity but also controlling the systematics related to this type of measurement and And in polar bear This is done in several ways. First of all, we can take the the specifications of our instrument and Simulate the uncertainty we have in our instrument So we can simulate the systematic and injected Injected in simulated time streams around those time streams in the in the science pipeline and see which b-mods spurious b-mod signal we get in our in our b-mod estimator as you can see here all those different Systematics are well below the b-mod signal level and also the statistical uncertainty of our measurement Of course this This takes well into account to the systematics the potential systematics that we already know but what about systematics so that we are not aware of and Most notably we would like to be able to assess or the quality of our data without looking at the final science product and this was a strict requirement in the polar bear collaboration because we adopted this blind policy in which we couldn't look at our final scientific product Before having accessed by other means the quality of the data and Performed the cut of our data So we perform the null test the null test unlike the simulated Test that we did before are done on the actual data but they are done by splitting the data into two halves and Subtracting them so that we expect to Find a signal in those data, which is compatible to the noise So with no feature due to spurious signals And this is done for several types of splitting of the data in order to probe different types of systematics all those system all those splitting of the data delivered something compatible with the noise we expect on those data and here you can see a Flat distribution which means that basically what we find all the distribution of these null test is compatible with the expected noise in those in those in those tests So what we learned after this? This season first of all we are very happy because we reached the sensitivity to get a detection and But we also learned that we control the systematic Systematics at this level of sensitivity and What we also learned that is that we are now close to the b-mod level Sensitivity and from other experiment we also learned that the foregrounds are an issue and And dominate the largest scales the one that were proven for by Thanks By experiment like bicep or plank So how do we plan the future accordingly? during this year Another telescope will be built close to the current one and the next year the the receiver here at KK in Japan will be delivered at the site and the focal plane of this new receiver will have six timer detectors and two frequency band Capability in order to begin to this disentangle between the signal and the foregrounds the year after in 2017 The next step of the polar bear experiment Will be the Simons array in which three telescopes will scan the sky at the same time the total number of detectors will exceed 22,000 and we will have also these other channel for better foreground rejection This is probably the most interesting slide for the audience. This is the Simons array sensitivity Also, including the presence of foregrounds. So how what we should expect from this experiment. So Using the multi-frequency high sensitivity of the Simons array and combining it with other foregrounds guards Like plunk and sea bass. We will be able to reject this Foreground level for the dust down to this level And this will enable to deliver as strong strong constraints on On the tensor to scalar ratio or the mass of the neutrinos. Here you can see the current Constrain and here are the forecast polar for cast Simon's array constraint and Here below I report I Are emphasized the numbers without accounting for foregrounds. So this is the the Target for the Simons array sensitivity, but taking into account realistically the The foregrounds. This is the level that we expect to achieve insensitivity and I emphasize how in five years the Polar bear collaboration is quickly improving its sensitivity moving towards tightly constrained the The physics related to inflation and other quest open questions in physics like the neutrino masses hierarchy and other more as exotic scenarios So this is the summary of this talk. I have shown you how the beam of measurement are a very active field nowadays and Polar bear is playing an important role in this quest and I've shown you the results of the first season in which we deliver the first measurement of Lensing beam of using CMB polarization alone and Validated this result with Cib cross correlation After the first season the task of continued scanning the sky and we already have those data and we are analyzing them right now About the future the high sensitivity of Polar bear 2 and Simons array will enable to probe both the lensing part of the power spectrum and the primordial power or part of the beam of beam odds and the multi-frequency nature with a wheel enable to Reject the foregrounds to high fidelity So that's all I wanted to say and thank you from everybody in this collaboration