Cognitive interference channel: achievable rate region and power allocation

Abstract

In this study, the authors consider a state-dependent two user interference channel. The two users sharing the spectrum are assumed to be cognitive and each user has a non-causal access to the signal from the other user. For this channel model, an achievable rate region is established for both discrete memoryless model and Gaussian channel. In particular, the achievable rate region is obtained by combining Han-Kobayashi rate splitting coding scheme, superposition coding, Gelfand-Pinsker coding scheme and zero-forcing dirty paper coding. Furthermore, the sum rate maximisation and the associated power allocation problem are studied, numerically and theoretically. The corresponding numerical examples show that the proposed combined coding scheme outperforms the existing schemes in the sense of achievable rate region. Moreover, the effectiveness of the optimal power allocation between the two cognitive nodes is also shown.

Performance analysis and power allocation for the unsaturated two‐user interference channel

“…In order to simplify our discussion, denote c i (q 1 , q 1 ) as the packet service rate of user i when the queue length is (q 1 , q 2 ). According to (12), D…”

“…Currently, there are a few works [11][12][13][14][15] on the performance in interference-limited networks. In [11][12][13], the system capacity has been analysed. The distribution of the received signal-to-inference-and-noise ratio when the transmit power is constant has been studied in [14].…”

“…The distribution of the received signal-to-inference-and-noise ratio when the transmit power is constant has been studied in [14]. Unlike [11][12][13][14], we focus on QoS performance of users.…”

On the capacity of the state‐dependent interference relay channel

“…10 The IRC was studied in many different scenarios. [11][12][13][14][15][16][17][18][19][20][21] For instance, the authors in Chang et al 13 investigated the IRC with a causal relay under strong and very strong interference regimes. In addition, Rini et al, Dytso et al, and Sridharan et al 14,15,18 studied the case in which both the users' signals are made available to the relay in a noncausal manner.…”

“…In their derivation, the main drawback is that the signals that are forwarded by the relay are not correlated with that transmitted from the sources. Al-qudah et al 20 derived an achievable rate region in which the two users are assumed to be dependent sources. Further, the state-dependent IRC (SIRC) was recently studied in Al-qudah and Rajan 19 in which the secondary user has a causal access to the primary signal and the two users' signals are also made causally available to the relay.…”