Interference Prediction in Wireless Networks Stochastic Geometry Meets Recursive Filtering
Abstract : This article proposes and evaluates a technique to predict the level of interference in wireless networks. We design a recursive predictor that estimates future interference values by filtering measured interference at a given location. The predictor’s parameterization is done offline by translating the autocorrelation of interference into an autoregressive moving average (ARMA) representation. This ARMA model is inserted into a steady-state Kalman filter enabling nodes to predict with low computational effort. Results show a good accuracy of predicted values versus true values for relevant time horizons. Although the predictor is parameterized for Poisson-distributed nodes, Rayleigh fading, and fixed message lengths, a sensitivity analysis shows that it also tends to work well in more general network scenarios. Numerical examples for underlay device-todevice communications, a common wireless sensor technology, and coexistence scenarios of Wi-Fi and LTE illustrate its broad applicability. The predictor can be applied as part of interference management to improve medium access, scheduling, and radio resource allocation.
? The existing state-of-the-art works on modeling and analysis of finite wireless networks, we can generally identify three streams of thoughts ordered in decreasing tractability and mathematical flexibility.
? It can generally be realized by adding new transmitters in the area of interest, which may be in the form of new BSs or distributed antennas from the existing ones.
? In addition to the previous considerations to be taken into account in the analysis of mmWave wireless networks, most of the existing literature based on SG and its inherent PP theory requires the inclusion of specific assumptions to ensure tractability.
? This is due to two main reasons. i) Existing link-level models for RISs are applicable to free-space channels and are, in general, formulated in terms of integrals or summations.
? The management of interference has always been a key issue in wireless systems.
? Simulations show that this predictor outperforms both basic predictors and predictors that consider only the channel dynamics (and disregard the impact of traffic).
? It is relevant to investigate the impact of transmissions allowed only if the sensed channel activity is below a given threshold from the transmission power.
? The impact of the design mismatch in terms of the inhomogeneous node distribution.
? Although interferers are not mobile in this scenario, interference is impacted by node mobility leading to similar effects.
• In, the authors proposed the use of the PHP to model a multi-cell D2D underlaid cellular network.
• Based on SG, the authors of proposed a multi-directional path loss model for fractal small cell networks, where the path loss exponent is modeled by i.i.d. random variables depending on the direction of signal propagation.
• In, a more generalized closed-form expression is proposed assuming an integer value of the path loss exponent.
• In, four approximation techniques are proposed based on the network operational regime.
• An alternative approach is proposed in to derive the average ergodic rate by considering general fading distributions and without necessarily going through the coverage probability expression.
? A setup with multiple wireless technologies coexisting in the same band contributes to more efficient use of scarce resources.
? The predictor is designed for predicting the interference arising from the Wi-Fi deployment alone and its performance is affected by the interference generated by the LTE system operating in the same band.
? The CCA modifies the Wi-Fi interference dynamics, as can be seen from the degraded performance of Wi-Fi using CCA with respect to Wi-Fi without CCA .
? The traces for Wi-Fi coexisting with LTE-U are below the trace for Wi-Fi using CCA alone, improving the predictor performance.
? Its performance analysis under matched and unmatched system conditions has demonstrated the prediction accuracy and robustness.
|