This paper develops a 3GPP-inspired design for the 1 in-band-full-duplex (IBFD) integrated access and backhaul (IAB) 2 networks in the frequency range 2 (FR2) band, which can 3 enhance the spectral efficiency (SE) and coverage while reducing 4 the latency. However, the self-interference (SI), which is usually 5 more than 100 dB higher than the signal-of-interest, becomes the 6 major bottleneck in developing these IBFD networks. We design 7 and analyze a subarray-based hybrid beamforming IBFD-IAB 8 system with the RF beamformers obtained via RF codebooks 9 given by a modified Linde-Buzo-Gray (LBG) algorithm. The 10 SI is canceled in three stages, where the first stage of antenna 11 isolation is assumed to be successfully deployed. The second stage 12 consists of the optical domain (OD)-based RF cancellation, where 13 cancelers are connected with the RF chain pairs. The third stage 14 is comprised of the digital cancellation via successive interference 15 cancellation followed by minimum mean-squared error baseband 16 receiver. Multiuser interference in the access link is canceled by 17 zero-forcing at the IAB-node transmitter. Simulations show that 18 under 400 MHz bandwidth, our proposed OD-based RF cancel-19 lation can achieve around 25 dB of cancellation with 100 taps. 20 Moreover, the higher the hardware impairment and channel 21 estimation error, the worse digital cancellation ability we can 22 obtain. 23 Index Terms-Wideband in-band-full-duplex millimeter wave 24 (FR2 band), subarray hybrid beamforming, integrated access and 25 backhaul, codebook design, self-interference cancellation. 26 I. INTRODUCTION 27 F REQUENCY range 2 (FR2) band (i.e., millimeter wave) 28 communications have been identified as the key technol-29 ogy for the beyond fifth-generation (5G) wireless communi-30 cations to provide much larger bandwidth, narrower beam, 31

In this paper, the problem of distributed content caching in a small-cell Base Stations (sBSs) wireless network that maximizes the cache hit performance is considered. Most of the existing works consider static demands, however, here, data at each sBS is considered to be correlated across time and sBSs. Federated learning (FL) based caching strategy is proposed which is assumed to be a weighted combination of past caching strategies of neighbouring base stations. A high probability generalization guarantees on the performance of the proposed federated caching strategy is derived. The theoretical guarantee provides following insights on obtaining the caching strategy: (i) run regret minimization at each sBS to obtain a sequence of caching strategies across time, and (ii) maximize an estimate of the bound to obtain a set of weights for the caching strategy which depends on the discrepancy. Also, theoretical guarantee on the performance of the least recently frequently used (LRFU) caching strategy is derived. Further, FL based heuristic caching algorithm is also proposed. Finally, it is shown through simulations using Movie Lens dataset that the proposed algorithm significantly outperforms the recent online learning algorithms.