Salient features of the pulsar emission are all consequences of a single property of the source of this radiation: the superluminal nature of the phase speeds with which the outer parts of the distribution pattern of the emitting electric current rotate. Here we outline the principles underlying the superluminal model of pulsars, on which this claim is based, and describe a ground-based experiment currently being built that will be capable of testing, inter alia, the validity of this model.

Conditions for the development of a two-stream instability in a pulsar magnetosphere are deduced from specific dispersion relations for plane wave perturbations, that depend on the distribution functions of the involved particles. Three different approaches are investigated.

Firstly, using relativistic one-dimensional Juttner Synge distribution functions appropriate to describe pulsar pair plasmas flows, we analytically derive the dispersion relation anew, precisely determining the dependence of its coefficients on the temperature, fluid velocity and associated Lorentz factor. We obtain modified frequencies for quasi-longitudinal waves and specific conditions for a two-stream instability to develop.

Secondly, the importance of two-stream instabilities is tested numerically on two different timescales, that concern stationary and non-stationary properties of pair plasma flows. The linear analysis involves Gaussian distribution functions of the momentum, factorized with functions that depend on the localisation of the different groups of particles, and shows results that agree with observed luminosities.

Finally as derived from fluid equations, the nonlinear evolution of such an instability process allows to associate the high level of radio radiation observed from pulsars with the existence of a lattice of radiating ‘Langmuir‘ soliton-like structures in a pulsar emission region. Actually, pair plasma particles follow the bundle of diverging magnetic field lines in the open magnetosphere and ‘Langmuir‘ soliton-like solutions, modified by magnetic field and density gradients, imply additional radiation.

Two-fluid effects on the radial 2D outflow of relativistic electron–positron plasma are considered. It is shown that when the longitudinal electric current is smaller than the Goldreich–Julian one, almost all electromagnetic energy is transformed into the energy of particles in the narrow boundary layer near the light surface |E| = |B|.

We investigate pulsar statistics based on theoretical model of radio emission of Ruderman-Sutherland (1975) and the revised spark model of Gil et al. (1999). In the present approach the intrinsic luminosity is no more given by observed relationship between Lr and P,. Our studies show that R-S & spark model can explain the statistical properties of radio pulsars reasonably well.

Several bright pulsars have position angle sweeps which are not consistent with a simple dipole field in the radio emission region. The data are now good enough to justify modelling of deviations in the field structure in these stars, and the effect of these deviations on the radio emission. As a first step, we consider the effect of quadrupoles, and of polar currents, on the observed polarization and emission signatures.

We show that in the inner magnetosphere of a pulsar mildly relativistic particles can flow out steadily under two assumptions: i) all vacuum fields are completely shielded by copious amounts of particles drawn out by thermal and field emission (which is likely as shown in Jessner et al, 1999) ii) particles emitted from the neutron star surface have weakly relativistic energies (βϒ ≈ 1). The results are consistent with the typical particle energies predicted in the radio emission model by Kunzl et al., 1999.

We present a new model for the coherent radio emission of pulsars that succeeds in reproducing observed luminosities at emission heights of 50-100 pulsar radii. Based on energy conservation and propagation effects we apply the concept of a free electron maser (FEM) to the creation of coherent radio emission. With the characteristic frequency ν ≃ γ2νpe (νpe denotes the plasma frequency and γ is the Lorentz factor of the emitting particles) we find that even for the lowest frequencies at which the highest degree of coherence is required only low energy particles (γ ≃ 10) with Goldreich-Julian-density are necessary. Because of its low energy budget we call it three-liter-pulsar.

The generation of radio emission from plasma waves excited by two-stream instability in pulsar magnetospheres is considered. Induced scattering transforms the excited longitudinal waves into waves that escape freely in the form of transverse electromagnetic waves. It is shown that the spectrum and the luminosity of the generated radio emission are compatible with those observed.

Relativistic plasma masers operating on the anomalous cyclotron-Cherenkov resonance ω − k||υ|| + ωB/ϒres = 0 and the Cherenkov-drift resonance ω − k||υ|| − kx/ud = 0, are capable of explaining the main observational characteristics of pulsar radio emission. Both electromagnetic instabilities are due to the interaction of the fast particles from the primary beam and from the tail of the secondary pairs distribution with the normal modes of a strongly magnetized one-dimensional electron-positron plasma. In a typical pulsar both resonances occur in the outer parts of magnetosphere at rres ≈ 109cm.

We show that under conditions of strong density modulation the effects of magnetospheric scintillations in diffractive and refractive regimes may be observable. The most distinctive feature of the magnetospheric scintillations is their independence on frequency. Results based on diffractive scattering due to small scale in homogeneities give a scattering angle that may be as large as 0.1 radians, and a typical decorrelation time of 10−8 seconds. Refractive scattering due to large scale inhomogeneities is also possible, with a typical angle of 10−3 radians and a correlation time of the order of 10−4 seconds. Temporal variation in the plasma density may also result in a delay time of the order of 10−4 seconds. The different scaling of the above quantities with frequency may allow one to distinguish the effects of propagation through a pulsar magnetosphere from the interstellar medium. In particular, we expect that the magnetospheric scintillations are relatively more important for nearby pulsars when observed at high frequencies.

Pulsar spectra have three characteristic parts: a low frequency turnover (the region of the maximum), a linear part with a constant spectral index, and a high frequency cut-off (after it the spectrum becomes considerably steeper) (Malofeev et al. 1994). These three parts can be described in the frame of the usual model of curvature radiation (Malov 1979; Ochelkov & Usov 1984; Kuz’min & Solov’ev 1986; Malov & Malofeev 1991). Kramer et al. (1997) have found a flattening in the spectra of 4 pulsars at frequencies above 30 GHz. It cannot be understood in such model. We propose the possible explanation for the unusual behaviour of these spectra.

New, self-consistent theory of coherent pulsar radio emission based on the sparking model of the polar gap first suggested by Ruderman & Sutherland (1975) and modified by Gil & Sendyk (1999), is proposed.

Formation of beam-plasma structure, a new non-linear object, in pulsar magnetosphere is considered.

The morphological classification of pulsar profiles is interpreted within the frame of the primordial hollow-cone emission model taking into account wave propagation in the magnetosphere. In this context the frequency evolution of the profiles as well as the difference in polarization properties of core and conal components can be explained naturally.

Many theoretical efforts were made to understand the core and conal emission identified from observations by Rankin (1983) and Lyne and Manchester (1988). One of them, named as inverse Compton scattering (ICS) model (Qiao & Lin 1998), has been proposed. It is found in the model that: there are central or ‘core’ emission beam, and one or two hollow conical emission beams; the different emission components are emitted at different heights; owing to different radiation components emitted from different height, the observed emission beams can be shifted from each other due to retardation and aberration effects; the sizes of emission components change with frequencies. Recent developments of the model include: simulations of pulse profiles at different frequencies; studying the basic polarization properties of inverse Compton scattering in strong magnetic fields; computing the polarizations and spectrum of core and cones. A new classification system was also proposed. The main results calculated from the model are consistent with the observations.

The radiative instability was considered in the context of pulsar radio emission by Goldreich & Keeley (1971) and rederived and extended by Asséo, Pellat, & Sol (1983) and by Asséo (1995). Their results can be generalised and reproduced by replacing the thin cylindrical ring of charged particles by an infinitesimally thin current carrying sheet in either planar or cylindrical geometry, suggesting that the local ring geometry may not be essential for the instability. It may be useful to further investigate the generalised dispersion equation given below.

First 1 emphasize the magnetospheric loop current system induced by the wind activity. Next, I address the formation of the field-aligned electric field. An interesting aspect is the reaction of the inner magnetosphere to an induced current. In the traditional outer gap, the current running through the gap is all carried by the particles created in the gap. However, if external current-carrying particles are added, the gap moves inward or outward; it can even appear near the star. It is suggested that the gaps appear at various altitudes, depending on the current distribution on field lines, the field line geometry, and the source of the current-carrying particles.

For definiteness of sign of charge and current direction, Ω · μ > 0 is assumed throughout.

We discuss plausible locations of radiating regions in the magnetospheres of pulsars and argue that the mechanisms of pulsar radiation at different frequencies are closely connected with the locations of the radiating regions, especially in the radio range. If the region that is responsible for the bulk of the non-thermal radiation at some frequency is localized in the pulsar magnetosphere then the nature of this radiation will be either determined or, at least, restricted significantly.

There are more than one thousand known radio pulsars, but only 9 of them have detected optical emission (Caraveo 1999). A part of optical emission can be caused by thermal radiation from the hot surface of the neutron star. We shall try to describe the non-thermal component (Kurt et al. 1980) on the base of the synchrotron mechanism, using one-dimensional distribution function of emitting electrons.

One of the possible reasons of observed pulsar emission is the cyclotron instability developed in an anisotropic plasma (Sagdeev & Shafranov 1960). To generate transversal (t) waves with the spectrum

the condition of the cyclotron resonance (Kazbegi et al. 1992)

must be fulfilled. Here is the Lorentz-factor of resonance particles, is the drift velocity of particles, ρ is the curvature radius of field lines.

The model provided that Geminga is an almost aligned pulsar, which is supported by the low-frequency observations. The model explains successfully most of the peculiarities of Geminga radio emission. In the frame of the same geometry we propose a mechanism for the recently detected optical pulsation from that pulsar.