We present results of H i absorption experiment done using the Giant Metrewave Radio Telescope (GMRT) towards 27 low- and intermediate-luminosity ($P_{\rm 1.4 GHz}\, \sim 10^{23}$-1026 W Hz−1) radio active galactic nuclei (AGNs), classified as either low excitation radio galaxies (LERGs) or high excitation radio galaxies (HERGs) and with WISE colour W2[4.6 μm]−W3[12 μm]> 2. We report H i absorption detection towards seven radio AGNs, six of which are new. Combined with other sources from literature classified as LERGs or HERGs, we confirm our earlier result that compact radio AGNs with WISE colour W2−W3 > 2 have higher detection rates compared to those with W2−W3 < 2. We find that H i absorption detection rate is higher for HERGs (37.0$^{+15.8}_{-11.5}$ per cent) compared to LERGs (22.0$^{+3.9}_{-3.4}$ per cent), mainly due to a larger fraction of HERGs being gas and dust rich with a younger stellar population compared to LERGs. However, for similar compact radio structures and host galaxies with WISE colours W2−W3 > 2, we don’t find any significant difference in detection rates of two types of AGNs implying detection of H i gas may not necessarily mean high excitation mode AGN. We further analysed the kinematics towards these sources. We find that while LERGs show a wide range in the shift of centroid velocities ($\sim \, -$479 to +356 km s−1) relative to the optical systemic velocity, most of the HERGs have centroid velocity shift less than 200 km s−1, possibly due to differences in jet-interstellar medium interaction.

Compact steep-spectrum (CSS) and peaked spectrum (PS) radio sources are compact, powerful radio sources. The multi-frequency observational properties and current theories are reviewed with emphasis on developments since the earlier review of O'Dea (1998). There are three main hypotheses for the nature of PS and CSS sources. (1) The PS sources might be very young radio galaxies which will evolve into CSS sources on their way to becoming large radio galaxies. (2) The PS and CSS sources might be compact because they are confined (and enhanced in radio power) by interaction with dense gas in their environments. (3) Alternately, the PS sources might be transient or intermittent sources. Each of these hypotheses may apply to individual objects. The relative number in each population will have significant implications for the radio galaxy paradigm. Proper motion studies over long time baselines have helped determine hotspot speeds for over three dozen sources and establish that these are young objects. Multifrequency polarization observations have demonstrated that many CSS/PS sources are embedded in a dense interstellar medium and vigorously interacting with it. The detection of emission line gas aligned with the radio source, and blue-shifted HI absorption and [OIII] emission lines indicates that AGN feedback is present in these objects -- possibly driven by the radio source. CSS/PS sources with evidence of episodic AGN over a large range of time-scales have been discussed. The review closes with a discussion of open questions and prospects for the future.