The DESY group is working on different Belle analysis and studies related to Belle II.

Precision Measurement of the Muon Pair Asymmetry in e+e- Annihilation at Belle

The Standard Model predicts interference effects between gamma and Z-boson exchange in muon pair production which cause a forward-backward asymmetry AFB even for energies well below the Z-pole. With the large available data samples of muon pairs collected at the Belle experiment during the last decade one expects a statistical uncertainty of about 1% at Belle and a statistical uncertainty as low as 0.1% at Belle II.

This precision requires a profund understanding of detector asymmetries, backgrounds and QED related corrections and respective systematic uncertainties. This energy-dependent asymmetry (AFB ~ -0.01 at 10.58 GeV) is proportional to the SM rho-parameter. This measurement is unique in the sense that it probes axial-axial operators and allows an extraction of the oblique parameter T that is independent of the oblique parameter S. A measurement of AFB with 0.1% precision would be complementary to world-best combined limits on T and sensitive to contact interaction scales competitive to proposed low energy parity violation experiments.

Studies of QCD-vacuum structure in electron-positron annihilation

One of the consequences of Heisenberg’s uncertainty relation is that the vacuum cannot be considered as space devoid of matter, but rather as a place where the basic laws of physics can be spontaneously violated over short periods of time allowing, e.g. the creation of particle-antiparticle pairs “out of nothing”. Any physical observable, both in the electroweak and strong sectors of the Standard Model, directly depends on the vacuum structure.

The QCD-vacuum structure will be studied through the process of hadronisation of quarks and antiquarks produced in e+e- annihilation at Belle and Belle II. The quarks cannot exist individually but combine with spontaneously created quarks and antiquarks from the vacuum and fragment into hadrons propagating through it. Therefore the process of hadronisation directly depends on the QCD-vacuum structure. The challenge lies in the fact that hadronisation itself is one of the unresolved problems in QCD.

At least three independent hadronisation mechanisms related to different manifestations of the vacuum structure are conceivable. Within the first approach, the formation of hadrons occurs in the perturbative QCD-vacuum where the study of hadronisation will deepen our knowledge on naive T-odd phenomenon. Within the second mechanism, the hadronisation occurs in CP-odd domains having the potential to provide an alternative view of the origin of fragmentation in the early stages of universe. In the third case, the quark- antiquark pair is produced in the non-perturbative structure of the QCD-vacuum which breaks the concept of factorisation.

The measurements will provide a wealth of information about the properties of hadronisation and thus will yield indirect information about the QCD-vacuum structure.

Neural network based reconstruction and analysis of the rare decay B->K*ll

Modern high energy physics has come to a point where the Standard Model of particle physics appears to be able to describe all phenomena and dynamics of the known subatomic particles. On the one hand, it holds against enduring testings and has proved its correctness for decades now. On the other hand, however, it has become clear that there must be physics beyond the Standard Model (SM) as it fails to describe fundamental parts of our universe. The ongoing testing did not show any signs for physics beyond the SM until now. Consequently, the room for "new physics" gets smaller day by day, and above all, it is becoming much more difficult to examine. One well-proven way to discover new physics is the search at higher energies, hoping to observe new particles or new behaviour of the known matter within the reachable energy scale. Other experiments aim to increase the precision on measurements in order to discover contributions of new physics in details.

The Belle experiment is designed as a so called "B Factory", allowing detailed studies of the B meson. The decay of the B meson is affected by all forces described in the Standard Model, so that many possible deviations can be examined.

We study flavour changing neutral currents, which are forbidden at tree level in the SM and are only allowed via higher-order diagrams.
This has a large effect on the branching fractions of these decays, so that we only expect one out of one million B mesons to decay into our desired sub particles. But the SM prediction for the decay amplitudes of B -> K l l can be calculated with small errors. Non-SM particles can contribute to the decay amplitude. Only a small deviation of the measured number of observed signal particles can end up in a large relative change to the theory, which makes these rare processes an ideal probe for "new physics".

The rare processes are however experimentally challenging because one has to handle millions of background B mesons in the Belle data set. We use artificial neural networks to learn the differences between signal and background events, helping us to find as much signal as possible.