Design of Coded ALOHA With ZigZag Decoder

Abstract

In this study, we investigated the effects of combining a zigzag decoder (ZD) with a coded ALOHA using the successive interference cancellation (SIC) technique to retrieve packets from collisions. We proposed zigzag decodable coded slotted ALOHA (ZDCSA) and enhanced-ZDCSA (E-ZDCSA) as a scheme which applies ZD before and after SIC, respectively. Furthermore, we derived the asymptotic analysis for the throughput and packet loss rate (PLR) performances of E-ZDCSA and validated its accuracy with the Monte Carlo simulations. Through numerical and asymptotic analyses, we showed that E-ZDCSA outperforms ZDCSA and the conventional coded ALOHA schemes in terms of throughput and PLR performances in most of the offered load regime. Moreover, we demonstrated that our proposed schemes outperform the conventional protocol in a practical scenario in which ZD cannot resolve all the collisions. INDEX TERMS Slotted ALOHA, successive interference cancellation, ZigZag decoding.

Collision Resolution for Random Access

“…3) Non-Unit Channel Gains With Imperfect CSI: For the ZigZag algorithm with imperfect CSI, since the channel estimates are not used in the decoupling process, the postprocessing system model is the same as the perfect CSI case, i.e., H𝑒𝑞 and H𝑒𝑞 are diagonal matrices consisting of the channel estimates and the channel estimation errors, respectively. So, using (10) and (11), the lower and upper bounds on the achievable sum-rate (per channel use) of ZigZag algorithms for single-user decoding in the presence of imperfect CSI are obtained as (30) and…”

“…The asymptotic throughput and packet loss rate of the E-ZDFA scheme are derived in [30] in a noise-free system model with unit channel gains. Unlike [29] and [30], we consider the effects of both noise and non-unit channel gains in this paper and determine information theoretic bounds on the achievable sum-rates, depicting that our proposed algorithms are asymptotically optimal. In addition, while [29] and [30] are able to resolve the collision of two users, we also propose a scheme to resolve the collision of more than two users.…”

Energy Harvesting Irregular Repetition ALOHA With Replica Concatenation

“…We also need to define the probability of successful concatenation with respect to an unknown replica for iteration updates, denoted by ζi (d), which can be written in terms of ζ i (d) as ζi (d) = ζi (d) pi . 14 We then specify ζ i and ζi as the expected success recovery probabilities for irregular repetition rates obtained by averaging ζ i (d) and ζi (d) over the user and replica degree probabilities, Λ d and λ d , as…”

“…The term (x)n is the Pochhammer symbol given by (x)n = Γ(x+n) Γ(x) . 14 We note that ζ i (d) is the total successful concatenation probability of a packet with d replicas, while ζi (d) is the conditional total successful concatenation probability on one of its unknown replicas. Therefore, we divide ζ i (d) by p i (previously defined as failure probability of a replica).…”

“…A more generic scheme, called coded slotted ALOHA (CSA) [10] replaces the repetition of packets with general linear block codes. ZigZag decoding is first proposed in [11] to solve the hidden terminal problem for wireless local networks, while it is also used for slotted ALOHA and carrier sensing multiple access [12], frameless ALOHA [13], and coded slotted ALOHA [14]. Many extensions on these advanced RA schemes for different scenarios are also studied in the literature (e.g., [15]- [22]).…”