Distributed ranking-based resource allocation for sporadic M2M communication

Abstract This work proposes a novel scheme for distributed ranking-based and contention-free resource allocation in large-scale machine-to-machine (M2M) communication networks. We partition network of N devices into disjoint clusters based on service type, assign to each cluster cluster-specific signature active members indicate their status. The are totally ordered some priori-known manner, which gives rise an ranking members. In order tackle complexity issues M2M networks with massive number devices, we propose using the framework compressed sensing (CS), mainly consists three phases: (i) full-duplex acquisition phase, transmit signatures simultaneously activation pattern is collected manner. (ii) base station detects per block sketching , allocates resources accordingly. (iii) Each device determines its accesses specific according position. By exploiting sparsity proposed formulated as CS support recovery problem particular binary block-sparse signal $$x\in {\mathbb{B}}^N$$ x ∈ B N – $$K_{B}$$ K in-block $$K_{I}$$ I over size d . Our analysis shows that efficiently reduces length $$\mathcal {O}(\max \{K_{B}\log N, K_{B}K_{I}\log d\})$$ O ( max { log , d } ) achieves less computational {O}(dK_{I}^{2}+\frac{N}{d}\log N)$$ 2 + compared standard algorithms. Moreover, numerical results suggest strong robustness under noisy conditions.