No abstract

With the continually increasing demand for higher data rate and link density as well as better network coverage, wireless networks, including 5G-enabled vehicular communication network (VCN), will be designed/deployed to cover more overlapping areas, experiencing more severe interferences. As a result, interference becomes a key impediment to the improvement of network performance, thus requiring a thorough investigation. There have been numerous interference management (IM) methods, of which interference alignment (IA) has been receiving significant attention in recent years. IA compresses the dimension of interference subspace by confining multiple interfering signals into a finite subspace, so that the desired signal subspace may be maximized. However, when IA is applied to the situation where multiple interferences are from the same source, the interference subspace cannot be compressed due to the fact that if these interfering components are aligned in the same direction at their unintended receiver (Rx), they will become indistinguishable at their desired Rx. To solve this problem, we propose Inside-Out Precoding (IOP). With IOP, multiple data streams that may cause interference to the other Rx are at first pre-processed at the interfering transmitter (Tx), by employing an inner-precoder which makes the streams spatially separable, and then, by exploiting interactions among wireless signals, multiple interfering components are treated as an effective interference to which an outer-precoder is applied, so that multiple interferences can be compressed into one dimension at the interfered Rx while making them distinguishable at their desired Rx. We present two IOP realizations -forward IOP (F-IOP) and backward IOP (B-IOP) -and propose a protocol to realize the synchronization of processing parameters at the interfering Tx and its intended Rx so that the Rx can adapt itself to the precoding strategy employed at the Tx side. Our in-depth analysis and simulation results have shown that the proposed IOP can effectively manage multiple interferences from the same source while guaranteeing the performance of transmission from the interfering Tx to its intended Rx.

An Adaptive tranSmission mechanism exploiting both interference loCality and the relationship between dEsired sigNal and inTerference (ASCENT) is proposed for uplink transmission in heterogeneous networks. The authors adopt both X channel and Z channel models according to which spatial signal processing is designed. In the X channel, the picocell base station (PBS) exploits information the macrocell base station (MBS) shares to cancel local interference, and cooperatively decodes the data carried by strong interference from a macro-user (MU), which is then fed to the MBS. As a result, a pico-user (PU) can transmit simultaneously with an MU on the same channel. In addition, adaptive reception is employed to achieve good tradeoff between interference suppression and the desired level of signal distortion. For the Z channel, the PU and the PBS adopt signal processing suitable for their own channel state. At a PBS, interference cancellation is adopted to eliminate disturbance from an MU via inter-base station collaboration. ASCENT is also extended to the case of multiple picocells. The authors' simulation results show that in X channel mode, the achievable uplink rate of an MU can be significantly enhanced. In the case of Z channel, the PU's rate is improved while guaranteeing the MU's data rate.