Feedback in Distributed Wireless Networks

In this talk, the use of feedback in distributed wireless networks is discussed using tools from information theory and game theory. In particular, the two-user Gaussian decentralized interference channel (DIC) is adopted as a study case. For this case, all rate-pairs that are achievable and stable when both transmitter-receiver pairs autonomously tune their own transmit/receive configurations to optimize their individual transmission rates (Nash region) are fully identified. The main message of this talk is threefold. First, it shows that thanks to the use of feedback, Pareto optimal operating points are achievable even in fully decentralized networks. Second, it shows that the existence of a feedback link is not a sufficient condition for enlarging the Nash region of the DIC with respect to the case in which feedback is not available. Indeed, when desired signals are stronger than interference, a larger NE region is observed only if the signal to noise ratios (SNRs) of the feedback links are higher than the signal to interference ratios (SIRs) and the interference to noise ratios (INRs) of the direct links. Conversely, when interference is stronger than the desired signals, a larger NE region is observed if the SNRs of the feedback links are higher than the SNRs of the direct links. Third, it is shown that previous results, namely the NE region of the linear deterministic DIC without feedback and the NE region of the linear deterministic DIC with perfect output feedback are obtained as special cases of the results presented in this talk.